diff --git a/.gitattributes b/.gitattributes
index 72b295fb431f2294183b2bacf217c52b0ce4a5e4..eb37209c874bce2f3e6e4d5e50aa1c4050416f37 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -397,3 +397,12 @@ Meta-Llama-3-8B-Instruct_int4_winogrande-routerbench-0shot_lr-0.0002_e-8_seq-512
 Qwen2-7B-Instruct_int4_medmcqa_full_con_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-89582-sd-1/checkpoint-11920/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen2-7B-Instruct_int4_medmcqa_full_con_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-89582-sd-1/checkpoint-23840/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen2-7B-Instruct_int4_medmcqa_full_con_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-89582-sd-1/checkpoint-35760/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6736da95acf74ac199158de2bcd2f45e279c6a48
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38b9d4be16ebe4fd6dec61f83710d425e98161ba60be90a89bdaeb9c11beb3ef
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..772f31cf2e416510d41cafab8c761d209cefcd08
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:deb96632b42907769e925b98a199f6fcdc0a7831f82dff2e4906d156ed0cb2a6
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4fb2328c3d69925f2df5bd45759191492150b200
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c03c76ae0896a0881fe52bd3b7a6e82c8932690a2e4f43c1a9dbbe48aaa334d6
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..855aac6548d56b320fc19c4b5d0eb6086d47ad28
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:056b2e3553bda71d84a70bc08025bb619086e3c0102977dd22d36e089f6403b6
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a6f216c1dabc2a6822f7dd858f323c56da9bb201
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cba4916bdb6b8f76363dde603abe1017bf6994aa96373586eaba8b95fa562bfd
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..2be701027ddef62ed06450c07b5aa2ca2b4edf6a
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/trainer_state.json
@@ -0,0 +1,7307 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 7.996895615056267,
+  "eval_steps": 10,
+  "global_step": 10304,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    },
+    {
+      "epoch": 3.0034924330616994,
+      "grad_norm": 0.5645653605461121,
+      "learning_rate": 0.0002,
+      "loss": 1.4983,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0112533954210323,
+      "grad_norm": 0.6457151174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0190143577803648,
+      "grad_norm": 0.583838164806366,
+      "learning_rate": 0.0002,
+      "loss": 1.3899,
+      "step": 3890
+    },
+    {
+      "epoch": 3.026775320139697,
+      "grad_norm": 0.6819260120391846,
+      "learning_rate": 0.0002,
+      "loss": 1.3258,
+      "step": 3900
+    },
+    {
+      "epoch": 3.03453628249903,
+      "grad_norm": 0.6692903637886047,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0422972448583625,
+      "grad_norm": 0.6101024746894836,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 3920
+    },
+    {
+      "epoch": 3.050058207217695,
+      "grad_norm": 0.7014093399047852,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3930
+    },
+    {
+      "epoch": 3.0578191695770274,
+      "grad_norm": 0.7380381226539612,
+      "learning_rate": 0.0002,
+      "loss": 1.3885,
+      "step": 3940
+    },
+    {
+      "epoch": 3.0655801319363603,
+      "grad_norm": 0.6607900857925415,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0733410942956927,
+      "grad_norm": 0.735263466835022,
+      "learning_rate": 0.0002,
+      "loss": 1.4293,
+      "step": 3960
+    },
+    {
+      "epoch": 3.081102056655025,
+      "grad_norm": 0.6788513660430908,
+      "learning_rate": 0.0002,
+      "loss": 1.3966,
+      "step": 3970
+    },
+    {
+      "epoch": 3.088863019014358,
+      "grad_norm": 0.6347652673721313,
+      "learning_rate": 0.0002,
+      "loss": 1.3435,
+      "step": 3980
+    },
+    {
+      "epoch": 3.0966239813736904,
+      "grad_norm": 0.7056642770767212,
+      "learning_rate": 0.0002,
+      "loss": 1.4518,
+      "step": 3990
+    },
+    {
+      "epoch": 3.104384943733023,
+      "grad_norm": 0.6387075185775757,
+      "learning_rate": 0.0002,
+      "loss": 1.4474,
+      "step": 4000
+    },
+    {
+      "epoch": 3.1121459060923553,
+      "grad_norm": 0.6701116561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.3833,
+      "step": 4010
+    },
+    {
+      "epoch": 3.119906868451688,
+      "grad_norm": 0.7558449506759644,
+      "learning_rate": 0.0002,
+      "loss": 1.404,
+      "step": 4020
+    },
+    {
+      "epoch": 3.1276678308110206,
+      "grad_norm": 0.6612881422042847,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 4030
+    },
+    {
+      "epoch": 3.135428793170353,
+      "grad_norm": 0.7474587559700012,
+      "learning_rate": 0.0002,
+      "loss": 1.439,
+      "step": 4040
+    },
+    {
+      "epoch": 3.1431897555296855,
+      "grad_norm": 0.7292373776435852,
+      "learning_rate": 0.0002,
+      "loss": 1.4616,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1509507178890184,
+      "grad_norm": 0.7432886958122253,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 4060
+    },
+    {
+      "epoch": 3.158711680248351,
+      "grad_norm": 0.6366098523139954,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1664726426076832,
+      "grad_norm": 0.6837611794471741,
+      "learning_rate": 0.0002,
+      "loss": 1.5044,
+      "step": 4080
+    },
+    {
+      "epoch": 3.174233604967016,
+      "grad_norm": 0.7194393277168274,
+      "learning_rate": 0.0002,
+      "loss": 1.4332,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1819945673263486,
+      "grad_norm": 0.6963607668876648,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 4100
+    },
+    {
+      "epoch": 3.189755529685681,
+      "grad_norm": 0.6404902935028076,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4110
+    },
+    {
+      "epoch": 3.1975164920450134,
+      "grad_norm": 0.7172070741653442,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 4120
+    },
+    {
+      "epoch": 3.2052774544043463,
+      "grad_norm": 0.6577759385108948,
+      "learning_rate": 0.0002,
+      "loss": 1.4658,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2130384167636787,
+      "grad_norm": 0.6658480167388916,
+      "learning_rate": 0.0002,
+      "loss": 1.4019,
+      "step": 4140
+    },
+    {
+      "epoch": 3.220799379123011,
+      "grad_norm": 0.6771699786186218,
+      "learning_rate": 0.0002,
+      "loss": 1.4348,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2285603414823436,
+      "grad_norm": 0.699035108089447,
+      "learning_rate": 0.0002,
+      "loss": 1.4736,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2363213038416765,
+      "grad_norm": 0.7218514680862427,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 4170
+    },
+    {
+      "epoch": 3.244082266201009,
+      "grad_norm": 0.6270631551742554,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 4180
+    },
+    {
+      "epoch": 3.2518432285603414,
+      "grad_norm": 0.6828921437263489,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2596041909196742,
+      "grad_norm": 0.6005498170852661,
+      "learning_rate": 0.0002,
+      "loss": 1.4663,
+      "step": 4200
+    },
+    {
+      "epoch": 3.2673651532790067,
+      "grad_norm": 0.6974790692329407,
+      "learning_rate": 0.0002,
+      "loss": 1.4798,
+      "step": 4210
+    },
+    {
+      "epoch": 3.275126115638339,
+      "grad_norm": 0.7269543409347534,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4220
+    },
+    {
+      "epoch": 3.2828870779976715,
+      "grad_norm": 0.6728787422180176,
+      "learning_rate": 0.0002,
+      "loss": 1.3848,
+      "step": 4230
+    },
+    {
+      "epoch": 3.2906480403570044,
+      "grad_norm": 0.676972508430481,
+      "learning_rate": 0.0002,
+      "loss": 1.4112,
+      "step": 4240
+    },
+    {
+      "epoch": 3.298409002716337,
+      "grad_norm": 0.748309314250946,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3061699650756693,
+      "grad_norm": 0.6976589560508728,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 4260
+    },
+    {
+      "epoch": 3.3139309274350017,
+      "grad_norm": 0.649780809879303,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3216918897943346,
+      "grad_norm": 0.6529902815818787,
+      "learning_rate": 0.0002,
+      "loss": 1.327,
+      "step": 4280
+    },
+    {
+      "epoch": 3.329452852153667,
+      "grad_norm": 0.9273163676261902,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 4290
+    },
+    {
+      "epoch": 3.3372138145129995,
+      "grad_norm": 0.717024028301239,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 4300
+    },
+    {
+      "epoch": 3.3449747768723324,
+      "grad_norm": 0.7914950251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.4441,
+      "step": 4310
+    },
+    {
+      "epoch": 3.352735739231665,
+      "grad_norm": 0.7133203148841858,
+      "learning_rate": 0.0002,
+      "loss": 1.432,
+      "step": 4320
+    },
+    {
+      "epoch": 3.3604967015909972,
+      "grad_norm": 0.7409568428993225,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3682576639503297,
+      "grad_norm": 0.6993981003761292,
+      "learning_rate": 0.0002,
+      "loss": 1.3992,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3760186263096625,
+      "grad_norm": 0.7114535570144653,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4350
+    },
+    {
+      "epoch": 3.383779588668995,
+      "grad_norm": 0.6790860295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.4227,
+      "step": 4360
+    },
+    {
+      "epoch": 3.3915405510283274,
+      "grad_norm": 0.6507849097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 4370
+    },
+    {
+      "epoch": 3.39930151338766,
+      "grad_norm": 0.5967804193496704,
+      "learning_rate": 0.0002,
+      "loss": 1.4559,
+      "step": 4380
+    },
+    {
+      "epoch": 3.4070624757469927,
+      "grad_norm": 0.6625847816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 4390
+    },
+    {
+      "epoch": 3.414823438106325,
+      "grad_norm": 0.6736508011817932,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4225844004656576,
+      "grad_norm": 0.7870860695838928,
+      "learning_rate": 0.0002,
+      "loss": 1.4363,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4303453628249905,
+      "grad_norm": 0.7205295562744141,
+      "learning_rate": 0.0002,
+      "loss": 1.4114,
+      "step": 4420
+    },
+    {
+      "epoch": 3.438106325184323,
+      "grad_norm": 0.6634634137153625,
+      "learning_rate": 0.0002,
+      "loss": 1.4131,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4458672875436553,
+      "grad_norm": 0.7562733292579651,
+      "learning_rate": 0.0002,
+      "loss": 1.4683,
+      "step": 4440
+    },
+    {
+      "epoch": 3.453628249902988,
+      "grad_norm": 0.6585879921913147,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4613892122623207,
+      "grad_norm": 0.6896792054176331,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 4460
+    },
+    {
+      "epoch": 3.469150174621653,
+      "grad_norm": 0.6520342230796814,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4769111369809855,
+      "grad_norm": 0.6760806441307068,
+      "learning_rate": 0.0002,
+      "loss": 1.3423,
+      "step": 4480
+    },
+    {
+      "epoch": 3.484672099340318,
+      "grad_norm": 0.7539774179458618,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 4490
+    },
+    {
+      "epoch": 3.492433061699651,
+      "grad_norm": 0.7409411668777466,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 4500
+    },
+    {
+      "epoch": 3.5001940240589833,
+      "grad_norm": 0.6876253485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.4069,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5079549864183157,
+      "grad_norm": 0.7028461694717407,
+      "learning_rate": 0.0002,
+      "loss": 1.4228,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5157159487776486,
+      "grad_norm": 0.8056529760360718,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4530
+    },
+    {
+      "epoch": 3.523476911136981,
+      "grad_norm": 0.711338996887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5312378734963135,
+      "grad_norm": 0.7343552708625793,
+      "learning_rate": 0.0002,
+      "loss": 1.5247,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5389988358556463,
+      "grad_norm": 0.745479941368103,
+      "learning_rate": 0.0002,
+      "loss": 1.4308,
+      "step": 4560
+    },
+    {
+      "epoch": 3.5467597982149788,
+      "grad_norm": 0.7582294940948486,
+      "learning_rate": 0.0002,
+      "loss": 1.4229,
+      "step": 4570
+    },
+    {
+      "epoch": 3.554520760574311,
+      "grad_norm": 0.6717444658279419,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4580
+    },
+    {
+      "epoch": 3.5622817229336436,
+      "grad_norm": 0.7417883276939392,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 4590
+    },
+    {
+      "epoch": 3.570042685292976,
+      "grad_norm": 0.6385737061500549,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4600
+    },
+    {
+      "epoch": 3.577803647652309,
+      "grad_norm": 0.716704249382019,
+      "learning_rate": 0.0002,
+      "loss": 1.3981,
+      "step": 4610
+    },
+    {
+      "epoch": 3.5855646100116414,
+      "grad_norm": 0.6948980093002319,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 4620
+    },
+    {
+      "epoch": 3.593325572370974,
+      "grad_norm": 0.6961140036582947,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6010865347303067,
+      "grad_norm": 0.7493122220039368,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 4640
+    },
+    {
+      "epoch": 3.608847497089639,
+      "grad_norm": 0.7431658506393433,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6166084594489716,
+      "grad_norm": 0.8353387713432312,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4660
+    },
+    {
+      "epoch": 3.6243694218083045,
+      "grad_norm": 0.7095612287521362,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 4670
+    },
+    {
+      "epoch": 3.632130384167637,
+      "grad_norm": 0.776620090007782,
+      "learning_rate": 0.0002,
+      "loss": 1.4003,
+      "step": 4680
+    },
+    {
+      "epoch": 3.6398913465269693,
+      "grad_norm": 0.7198925018310547,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6476523088863018,
+      "grad_norm": 0.8238834738731384,
+      "learning_rate": 0.0002,
+      "loss": 1.4543,
+      "step": 4700
+    },
+    {
+      "epoch": 3.655413271245634,
+      "grad_norm": 0.6804245710372925,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4710
+    },
+    {
+      "epoch": 3.663174233604967,
+      "grad_norm": 0.8444845676422119,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 4720
+    },
+    {
+      "epoch": 3.6709351959642995,
+      "grad_norm": 0.743797779083252,
+      "learning_rate": 0.0002,
+      "loss": 1.3825,
+      "step": 4730
+    },
+    {
+      "epoch": 3.678696158323632,
+      "grad_norm": 0.8994188904762268,
+      "learning_rate": 0.0002,
+      "loss": 1.4213,
+      "step": 4740
+    },
+    {
+      "epoch": 3.686457120682965,
+      "grad_norm": 0.75416100025177,
+      "learning_rate": 0.0002,
+      "loss": 1.4281,
+      "step": 4750
+    },
+    {
+      "epoch": 3.6942180830422973,
+      "grad_norm": 0.6499266028404236,
+      "learning_rate": 0.0002,
+      "loss": 1.4154,
+      "step": 4760
+    },
+    {
+      "epoch": 3.7019790454016297,
+      "grad_norm": 0.7246791124343872,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7097400077609626,
+      "grad_norm": 0.7831124067306519,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 4780
+    },
+    {
+      "epoch": 3.717500970120295,
+      "grad_norm": 0.7130028009414673,
+      "learning_rate": 0.0002,
+      "loss": 1.3933,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7252619324796274,
+      "grad_norm": 0.7501602172851562,
+      "learning_rate": 0.0002,
+      "loss": 1.4632,
+      "step": 4800
+    },
+    {
+      "epoch": 3.73302289483896,
+      "grad_norm": 0.6980932950973511,
+      "learning_rate": 0.0002,
+      "loss": 1.4985,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7407838571982923,
+      "grad_norm": 0.8050530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.4517,
+      "step": 4820
+    },
+    {
+      "epoch": 3.748544819557625,
+      "grad_norm": 0.6385579705238342,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7563057819169576,
+      "grad_norm": 0.6664714813232422,
+      "learning_rate": 0.0002,
+      "loss": 1.5281,
+      "step": 4840
+    },
+    {
+      "epoch": 3.76406674427629,
+      "grad_norm": 0.7125676274299622,
+      "learning_rate": 0.0002,
+      "loss": 1.4443,
+      "step": 4850
+    },
+    {
+      "epoch": 3.771827706635623,
+      "grad_norm": 0.7231866717338562,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4860
+    },
+    {
+      "epoch": 3.7795886689949554,
+      "grad_norm": 0.6917183995246887,
+      "learning_rate": 0.0002,
+      "loss": 1.4446,
+      "step": 4870
+    },
+    {
+      "epoch": 3.787349631354288,
+      "grad_norm": 0.665037989616394,
+      "learning_rate": 0.0002,
+      "loss": 1.4369,
+      "step": 4880
+    },
+    {
+      "epoch": 3.7951105937136207,
+      "grad_norm": 0.5837726593017578,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4890
+    },
+    {
+      "epoch": 3.802871556072953,
+      "grad_norm": 0.6366701722145081,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8106325184322856,
+      "grad_norm": 0.7082223892211914,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 4910
+    },
+    {
+      "epoch": 3.818393480791618,
+      "grad_norm": 0.8101672530174255,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4920
+    },
+    {
+      "epoch": 3.826154443150951,
+      "grad_norm": 0.7516148090362549,
+      "learning_rate": 0.0002,
+      "loss": 1.3659,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8339154055102833,
+      "grad_norm": 0.7928489446640015,
+      "learning_rate": 0.0002,
+      "loss": 1.3909,
+      "step": 4940
+    },
+    {
+      "epoch": 3.8416763678696157,
+      "grad_norm": 0.6892234683036804,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 4950
+    },
+    {
+      "epoch": 3.849437330228948,
+      "grad_norm": 0.6381304264068604,
+      "learning_rate": 0.0002,
+      "loss": 1.5024,
+      "step": 4960
+    },
+    {
+      "epoch": 3.857198292588281,
+      "grad_norm": 0.8068831562995911,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8649592549476135,
+      "grad_norm": 0.7289869785308838,
+      "learning_rate": 0.0002,
+      "loss": 1.45,
+      "step": 4980
+    },
+    {
+      "epoch": 3.872720217306946,
+      "grad_norm": 0.7278549075126648,
+      "learning_rate": 0.0002,
+      "loss": 1.398,
+      "step": 4990
+    },
+    {
+      "epoch": 3.880481179666279,
+      "grad_norm": 0.7324236631393433,
+      "learning_rate": 0.0002,
+      "loss": 1.4442,
+      "step": 5000
+    },
+    {
+      "epoch": 3.8882421420256112,
+      "grad_norm": 0.6759871244430542,
+      "learning_rate": 0.0002,
+      "loss": 1.4511,
+      "step": 5010
+    },
+    {
+      "epoch": 3.8960031043849437,
+      "grad_norm": 0.8159207701683044,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 5020
+    },
+    {
+      "epoch": 3.9037640667442766,
+      "grad_norm": 0.6536211967468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5030
+    },
+    {
+      "epoch": 3.911525029103609,
+      "grad_norm": 0.6827932000160217,
+      "learning_rate": 0.0002,
+      "loss": 1.4335,
+      "step": 5040
+    },
+    {
+      "epoch": 3.9192859914629414,
+      "grad_norm": 0.6688340306282043,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 5050
+    },
+    {
+      "epoch": 3.927046953822274,
+      "grad_norm": 0.6385695934295654,
+      "learning_rate": 0.0002,
+      "loss": 1.4099,
+      "step": 5060
+    },
+    {
+      "epoch": 3.9348079161816063,
+      "grad_norm": 0.6975107192993164,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 5070
+    },
+    {
+      "epoch": 3.942568878540939,
+      "grad_norm": 0.6684112548828125,
+      "learning_rate": 0.0002,
+      "loss": 1.4893,
+      "step": 5080
+    },
+    {
+      "epoch": 3.9503298409002716,
+      "grad_norm": 0.8349628448486328,
+      "learning_rate": 0.0002,
+      "loss": 1.4732,
+      "step": 5090
+    },
+    {
+      "epoch": 3.958090803259604,
+      "grad_norm": 0.7146425843238831,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 5100
+    },
+    {
+      "epoch": 3.965851765618937,
+      "grad_norm": 0.6555036902427673,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9736127279782694,
+      "grad_norm": 0.7037415504455566,
+      "learning_rate": 0.0002,
+      "loss": 1.4274,
+      "step": 5120
+    },
+    {
+      "epoch": 3.981373690337602,
+      "grad_norm": 0.7235575914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5130
+    },
+    {
+      "epoch": 3.9891346526969347,
+      "grad_norm": 0.7092325687408447,
+      "learning_rate": 0.0002,
+      "loss": 1.4455,
+      "step": 5140
+    },
+    {
+      "epoch": 3.996895615056267,
+      "grad_norm": 0.7490319609642029,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 5150
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9131355285644531,
+      "eval_runtime": 105.5778,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 5154
+    },
+    {
+      "epoch": 4.0046565774155995,
+      "grad_norm": 0.7075854539871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 5160
+    },
+    {
+      "epoch": 4.012417539774932,
+      "grad_norm": 0.9466007351875305,
+      "learning_rate": 0.0002,
+      "loss": 1.209,
+      "step": 5170
+    },
+    {
+      "epoch": 4.020178502134264,
+      "grad_norm": 1.0297044515609741,
+      "learning_rate": 0.0002,
+      "loss": 1.2567,
+      "step": 5180
+    },
+    {
+      "epoch": 4.027939464493597,
+      "grad_norm": 0.7765059471130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5190
+    },
+    {
+      "epoch": 4.03570042685293,
+      "grad_norm": 0.995760977268219,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 5200
+    },
+    {
+      "epoch": 4.043461389212262,
+      "grad_norm": 0.8663829565048218,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 5210
+    },
+    {
+      "epoch": 4.051222351571595,
+      "grad_norm": 1.0660825967788696,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 5220
+    },
+    {
+      "epoch": 4.058983313930927,
+      "grad_norm": 0.9858174920082092,
+      "learning_rate": 0.0002,
+      "loss": 1.1676,
+      "step": 5230
+    },
+    {
+      "epoch": 4.06674427629026,
+      "grad_norm": 0.8911338448524475,
+      "learning_rate": 0.0002,
+      "loss": 1.2448,
+      "step": 5240
+    },
+    {
+      "epoch": 4.074505238649593,
+      "grad_norm": 1.0848394632339478,
+      "learning_rate": 0.0002,
+      "loss": 1.1858,
+      "step": 5250
+    },
+    {
+      "epoch": 4.082266201008925,
+      "grad_norm": 1.0849905014038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5260
+    },
+    {
+      "epoch": 4.090027163368258,
+      "grad_norm": 1.0497841835021973,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 5270
+    },
+    {
+      "epoch": 4.0977881257275905,
+      "grad_norm": 0.8943053483963013,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 5280
+    },
+    {
+      "epoch": 4.1055490880869225,
+      "grad_norm": 0.8432527184486389,
+      "learning_rate": 0.0002,
+      "loss": 1.1923,
+      "step": 5290
+    },
+    {
+      "epoch": 4.113310050446255,
+      "grad_norm": 0.9690414667129517,
+      "learning_rate": 0.0002,
+      "loss": 1.1634,
+      "step": 5300
+    },
+    {
+      "epoch": 4.121071012805588,
+      "grad_norm": 0.7790773510932922,
+      "learning_rate": 0.0002,
+      "loss": 1.3019,
+      "step": 5310
+    },
+    {
+      "epoch": 4.12883197516492,
+      "grad_norm": 0.9289211630821228,
+      "learning_rate": 0.0002,
+      "loss": 1.1806,
+      "step": 5320
+    },
+    {
+      "epoch": 4.136592937524253,
+      "grad_norm": 1.0785125494003296,
+      "learning_rate": 0.0002,
+      "loss": 1.1458,
+      "step": 5330
+    },
+    {
+      "epoch": 4.144353899883585,
+      "grad_norm": 0.8559591770172119,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5340
+    },
+    {
+      "epoch": 4.152114862242918,
+      "grad_norm": 0.9405956268310547,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5350
+    },
+    {
+      "epoch": 4.159875824602251,
+      "grad_norm": 0.9942827820777893,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 5360
+    },
+    {
+      "epoch": 4.167636786961583,
+      "grad_norm": 0.9141933917999268,
+      "learning_rate": 0.0002,
+      "loss": 1.1659,
+      "step": 5370
+    },
+    {
+      "epoch": 4.175397749320916,
+      "grad_norm": 0.8206015229225159,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 5380
+    },
+    {
+      "epoch": 4.183158711680249,
+      "grad_norm": 0.9340888857841492,
+      "learning_rate": 0.0002,
+      "loss": 1.2778,
+      "step": 5390
+    },
+    {
+      "epoch": 4.190919674039581,
+      "grad_norm": 1.2122114896774292,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 5400
+    },
+    {
+      "epoch": 4.1986806363989135,
+      "grad_norm": 1.0661298036575317,
+      "learning_rate": 0.0002,
+      "loss": 1.2371,
+      "step": 5410
+    },
+    {
+      "epoch": 4.206441598758246,
+      "grad_norm": 0.9372861385345459,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 5420
+    },
+    {
+      "epoch": 4.214202561117578,
+      "grad_norm": 0.894012987613678,
+      "learning_rate": 0.0002,
+      "loss": 1.2653,
+      "step": 5430
+    },
+    {
+      "epoch": 4.221963523476911,
+      "grad_norm": 1.0647753477096558,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5440
+    },
+    {
+      "epoch": 4.229724485836243,
+      "grad_norm": 0.989179790019989,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 5450
+    },
+    {
+      "epoch": 4.237485448195576,
+      "grad_norm": 1.1601181030273438,
+      "learning_rate": 0.0002,
+      "loss": 1.2715,
+      "step": 5460
+    },
+    {
+      "epoch": 4.245246410554909,
+      "grad_norm": 0.9395585656166077,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 5470
+    },
+    {
+      "epoch": 4.253007372914241,
+      "grad_norm": 0.9527766108512878,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 5480
+    },
+    {
+      "epoch": 4.260768335273574,
+      "grad_norm": 1.0319520235061646,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 5490
+    },
+    {
+      "epoch": 4.268529297632907,
+      "grad_norm": 0.8659824728965759,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 5500
+    },
+    {
+      "epoch": 4.276290259992239,
+      "grad_norm": 1.099211573600769,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 5510
+    },
+    {
+      "epoch": 4.284051222351572,
+      "grad_norm": 0.9363361597061157,
+      "learning_rate": 0.0002,
+      "loss": 1.2508,
+      "step": 5520
+    },
+    {
+      "epoch": 4.2918121847109045,
+      "grad_norm": 0.8437647223472595,
+      "learning_rate": 0.0002,
+      "loss": 1.189,
+      "step": 5530
+    },
+    {
+      "epoch": 4.2995731470702365,
+      "grad_norm": 0.9181258678436279,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5540
+    },
+    {
+      "epoch": 4.307334109429569,
+      "grad_norm": 0.9059357643127441,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 5550
+    },
+    {
+      "epoch": 4.315095071788901,
+      "grad_norm": 0.9337241649627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2189,
+      "step": 5560
+    },
+    {
+      "epoch": 4.322856034148234,
+      "grad_norm": 0.9428889155387878,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 5570
+    },
+    {
+      "epoch": 4.330616996507567,
+      "grad_norm": 1.003589153289795,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 5580
+    },
+    {
+      "epoch": 4.338377958866899,
+      "grad_norm": 1.1249268054962158,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 5590
+    },
+    {
+      "epoch": 4.346138921226232,
+      "grad_norm": 0.8623469471931458,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 5600
+    },
+    {
+      "epoch": 4.353899883585565,
+      "grad_norm": 1.1389174461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5610
+    },
+    {
+      "epoch": 4.361660845944897,
+      "grad_norm": 1.0136264562606812,
+      "learning_rate": 0.0002,
+      "loss": 1.2245,
+      "step": 5620
+    },
+    {
+      "epoch": 4.36942180830423,
+      "grad_norm": 0.9567070603370667,
+      "learning_rate": 0.0002,
+      "loss": 1.3473,
+      "step": 5630
+    },
+    {
+      "epoch": 4.377182770663563,
+      "grad_norm": 1.0592148303985596,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 5640
+    },
+    {
+      "epoch": 4.384943733022895,
+      "grad_norm": 1.0110485553741455,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5650
+    },
+    {
+      "epoch": 4.3927046953822275,
+      "grad_norm": 0.9914907217025757,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5660
+    },
+    {
+      "epoch": 4.4004656577415595,
+      "grad_norm": 0.9447247982025146,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 5670
+    },
+    {
+      "epoch": 4.408226620100892,
+      "grad_norm": 0.9644378423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 5680
+    },
+    {
+      "epoch": 4.415987582460225,
+      "grad_norm": 0.920676589012146,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 5690
+    },
+    {
+      "epoch": 4.423748544819557,
+      "grad_norm": 1.060570478439331,
+      "learning_rate": 0.0002,
+      "loss": 1.2792,
+      "step": 5700
+    },
+    {
+      "epoch": 4.43150950717889,
+      "grad_norm": 0.8857738971710205,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5710
+    },
+    {
+      "epoch": 4.439270469538223,
+      "grad_norm": 1.0536398887634277,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 5720
+    },
+    {
+      "epoch": 4.447031431897555,
+      "grad_norm": 0.990847110748291,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 5730
+    },
+    {
+      "epoch": 4.454792394256888,
+      "grad_norm": 0.9692499041557312,
+      "learning_rate": 0.0002,
+      "loss": 1.2469,
+      "step": 5740
+    },
+    {
+      "epoch": 4.462553356616221,
+      "grad_norm": 1.0376402139663696,
+      "learning_rate": 0.0002,
+      "loss": 1.2269,
+      "step": 5750
+    },
+    {
+      "epoch": 4.470314318975553,
+      "grad_norm": 1.3863259553909302,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 5760
+    },
+    {
+      "epoch": 4.478075281334886,
+      "grad_norm": 0.978379487991333,
+      "learning_rate": 0.0002,
+      "loss": 1.2591,
+      "step": 5770
+    },
+    {
+      "epoch": 4.485836243694218,
+      "grad_norm": 1.0973085165023804,
+      "learning_rate": 0.0002,
+      "loss": 1.2729,
+      "step": 5780
+    },
+    {
+      "epoch": 4.4935972060535505,
+      "grad_norm": 1.057006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5790
+    },
+    {
+      "epoch": 4.501358168412883,
+      "grad_norm": 0.9247729182243347,
+      "learning_rate": 0.0002,
+      "loss": 1.2476,
+      "step": 5800
+    },
+    {
+      "epoch": 4.509119130772215,
+      "grad_norm": 1.0447787046432495,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 5810
+    },
+    {
+      "epoch": 4.516880093131548,
+      "grad_norm": 1.1930429935455322,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 5820
+    },
+    {
+      "epoch": 4.524641055490881,
+      "grad_norm": 0.9867590069770813,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5830
+    },
+    {
+      "epoch": 4.532402017850213,
+      "grad_norm": 0.9591100215911865,
+      "learning_rate": 0.0002,
+      "loss": 1.2766,
+      "step": 5840
+    },
+    {
+      "epoch": 4.540162980209546,
+      "grad_norm": 0.9950753450393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 5850
+    },
+    {
+      "epoch": 4.547923942568879,
+      "grad_norm": 1.0087506771087646,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 5860
+    },
+    {
+      "epoch": 4.555684904928211,
+      "grad_norm": 1.0934417247772217,
+      "learning_rate": 0.0002,
+      "loss": 1.3165,
+      "step": 5870
+    },
+    {
+      "epoch": 4.563445867287544,
+      "grad_norm": 1.107987403869629,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 5880
+    },
+    {
+      "epoch": 4.571206829646876,
+      "grad_norm": 0.9147276878356934,
+      "learning_rate": 0.0002,
+      "loss": 1.2184,
+      "step": 5890
+    },
+    {
+      "epoch": 4.578967792006209,
+      "grad_norm": 1.036780595779419,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 5900
+    },
+    {
+      "epoch": 4.5867287543655415,
+      "grad_norm": 0.9284719824790955,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 5910
+    },
+    {
+      "epoch": 4.5944897167248735,
+      "grad_norm": 0.9141898155212402,
+      "learning_rate": 0.0002,
+      "loss": 1.3693,
+      "step": 5920
+    },
+    {
+      "epoch": 4.602250679084206,
+      "grad_norm": 1.0447357892990112,
+      "learning_rate": 0.0002,
+      "loss": 1.2319,
+      "step": 5930
+    },
+    {
+      "epoch": 4.610011641443539,
+      "grad_norm": 0.9309114217758179,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 5940
+    },
+    {
+      "epoch": 4.617772603802871,
+      "grad_norm": 1.2986129522323608,
+      "learning_rate": 0.0002,
+      "loss": 1.2827,
+      "step": 5950
+    },
+    {
+      "epoch": 4.625533566162204,
+      "grad_norm": 0.9221704602241516,
+      "learning_rate": 0.0002,
+      "loss": 1.312,
+      "step": 5960
+    },
+    {
+      "epoch": 4.633294528521537,
+      "grad_norm": 0.9228187799453735,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 5970
+    },
+    {
+      "epoch": 4.641055490880869,
+      "grad_norm": 0.9483116269111633,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 5980
+    },
+    {
+      "epoch": 4.648816453240202,
+      "grad_norm": 1.0218974351882935,
+      "learning_rate": 0.0002,
+      "loss": 1.3437,
+      "step": 5990
+    },
+    {
+      "epoch": 4.656577415599534,
+      "grad_norm": 0.9764600396156311,
+      "learning_rate": 0.0002,
+      "loss": 1.3085,
+      "step": 6000
+    },
+    {
+      "epoch": 4.664338377958867,
+      "grad_norm": 0.9115710258483887,
+      "learning_rate": 0.0002,
+      "loss": 1.197,
+      "step": 6010
+    },
+    {
+      "epoch": 4.6720993403182,
+      "grad_norm": 0.9245651364326477,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 6020
+    },
+    {
+      "epoch": 4.6798603026775325,
+      "grad_norm": 0.9686311483383179,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 6030
+    },
+    {
+      "epoch": 4.6876212650368645,
+      "grad_norm": 1.1807392835617065,
+      "learning_rate": 0.0002,
+      "loss": 1.2702,
+      "step": 6040
+    },
+    {
+      "epoch": 4.695382227396197,
+      "grad_norm": 1.0358641147613525,
+      "learning_rate": 0.0002,
+      "loss": 1.328,
+      "step": 6050
+    },
+    {
+      "epoch": 4.703143189755529,
+      "grad_norm": 0.987332284450531,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6060
+    },
+    {
+      "epoch": 4.710904152114862,
+      "grad_norm": 1.0526494979858398,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6070
+    },
+    {
+      "epoch": 4.718665114474195,
+      "grad_norm": 1.0276758670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.2246,
+      "step": 6080
+    },
+    {
+      "epoch": 4.726426076833527,
+      "grad_norm": 0.9904406666755676,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6090
+    },
+    {
+      "epoch": 4.73418703919286,
+      "grad_norm": 1.0084882974624634,
+      "learning_rate": 0.0002,
+      "loss": 1.2797,
+      "step": 6100
+    },
+    {
+      "epoch": 4.741948001552192,
+      "grad_norm": 0.8646450638771057,
+      "learning_rate": 0.0002,
+      "loss": 1.2656,
+      "step": 6110
+    },
+    {
+      "epoch": 4.749708963911525,
+      "grad_norm": 0.9233377575874329,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 6120
+    },
+    {
+      "epoch": 4.757469926270858,
+      "grad_norm": 0.9675140976905823,
+      "learning_rate": 0.0002,
+      "loss": 1.2642,
+      "step": 6130
+    },
+    {
+      "epoch": 4.765230888630191,
+      "grad_norm": 0.9639796018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6140
+    },
+    {
+      "epoch": 4.772991850989523,
+      "grad_norm": 0.925199568271637,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 6150
+    },
+    {
+      "epoch": 4.7807528133488555,
+      "grad_norm": 1.050901174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 6160
+    },
+    {
+      "epoch": 4.7885137757081875,
+      "grad_norm": 0.8920623660087585,
+      "learning_rate": 0.0002,
+      "loss": 1.301,
+      "step": 6170
+    },
+    {
+      "epoch": 4.79627473806752,
+      "grad_norm": 0.8964757919311523,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6180
+    },
+    {
+      "epoch": 4.804035700426853,
+      "grad_norm": 1.0839070081710815,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 6190
+    },
+    {
+      "epoch": 4.811796662786185,
+      "grad_norm": 0.8809942007064819,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 6200
+    },
+    {
+      "epoch": 4.819557625145518,
+      "grad_norm": 1.0216195583343506,
+      "learning_rate": 0.0002,
+      "loss": 1.321,
+      "step": 6210
+    },
+    {
+      "epoch": 4.827318587504851,
+      "grad_norm": 0.892005980014801,
+      "learning_rate": 0.0002,
+      "loss": 1.3033,
+      "step": 6220
+    },
+    {
+      "epoch": 4.835079549864183,
+      "grad_norm": 0.9957166910171509,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6230
+    },
+    {
+      "epoch": 4.842840512223516,
+      "grad_norm": 0.9720533490180969,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 6240
+    },
+    {
+      "epoch": 4.850601474582849,
+      "grad_norm": 0.9336182475090027,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 6250
+    },
+    {
+      "epoch": 4.858362436942181,
+      "grad_norm": 1.2611457109451294,
+      "learning_rate": 0.0002,
+      "loss": 1.3136,
+      "step": 6260
+    },
+    {
+      "epoch": 4.866123399301514,
+      "grad_norm": 0.8927203416824341,
+      "learning_rate": 0.0002,
+      "loss": 1.2234,
+      "step": 6270
+    },
+    {
+      "epoch": 4.873884361660846,
+      "grad_norm": 0.9706710577011108,
+      "learning_rate": 0.0002,
+      "loss": 1.3463,
+      "step": 6280
+    },
+    {
+      "epoch": 4.8816453240201785,
+      "grad_norm": 1.1461690664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 6290
+    },
+    {
+      "epoch": 4.889406286379511,
+      "grad_norm": 0.9930381178855896,
+      "learning_rate": 0.0002,
+      "loss": 1.2566,
+      "step": 6300
+    },
+    {
+      "epoch": 4.897167248738843,
+      "grad_norm": 0.91451096534729,
+      "learning_rate": 0.0002,
+      "loss": 1.2568,
+      "step": 6310
+    },
+    {
+      "epoch": 4.904928211098176,
+      "grad_norm": 1.0319571495056152,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 6320
+    },
+    {
+      "epoch": 4.912689173457509,
+      "grad_norm": 0.990140438079834,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6330
+    },
+    {
+      "epoch": 4.920450135816841,
+      "grad_norm": 1.2466117143630981,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 6340
+    },
+    {
+      "epoch": 4.928211098176174,
+      "grad_norm": 1.0316979885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 6350
+    },
+    {
+      "epoch": 4.935972060535507,
+      "grad_norm": 1.0643759965896606,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 6360
+    },
+    {
+      "epoch": 4.943733022894839,
+      "grad_norm": 0.9703279733657837,
+      "learning_rate": 0.0002,
+      "loss": 1.2559,
+      "step": 6370
+    },
+    {
+      "epoch": 4.951493985254172,
+      "grad_norm": 0.9767927527427673,
+      "learning_rate": 0.0002,
+      "loss": 1.2155,
+      "step": 6380
+    },
+    {
+      "epoch": 4.959254947613504,
+      "grad_norm": 0.960854172706604,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 6390
+    },
+    {
+      "epoch": 4.967015909972837,
+      "grad_norm": 0.9922910332679749,
+      "learning_rate": 0.0002,
+      "loss": 1.3314,
+      "step": 6400
+    },
+    {
+      "epoch": 4.9747768723321695,
+      "grad_norm": 0.956470787525177,
+      "learning_rate": 0.0002,
+      "loss": 1.3018,
+      "step": 6410
+    },
+    {
+      "epoch": 4.9825378346915015,
+      "grad_norm": 0.9637242555618286,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 6420
+    },
+    {
+      "epoch": 4.990298797050834,
+      "grad_norm": 1.0855202674865723,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 6430
+    },
+    {
+      "epoch": 4.998059759410167,
+      "grad_norm": 0.9655316472053528,
+      "learning_rate": 0.0002,
+      "loss": 1.3015,
+      "step": 6440
+    },
+    {
+      "epoch": 4.9996119518820334,
+      "eval_loss": 2.0410802364349365,
+      "eval_runtime": 113.04,
+      "eval_samples_per_second": 4.485,
+      "eval_steps_per_second": 0.566,
+      "step": 6442
+    },
+    {
+      "epoch": 5.005820721769499,
+      "grad_norm": 1.1676199436187744,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 6450
+    },
+    {
+      "epoch": 5.013581684128832,
+      "grad_norm": 1.4317965507507324,
+      "learning_rate": 0.0002,
+      "loss": 1.041,
+      "step": 6460
+    },
+    {
+      "epoch": 5.021342646488165,
+      "grad_norm": 1.460443377494812,
+      "learning_rate": 0.0002,
+      "loss": 0.9546,
+      "step": 6470
+    },
+    {
+      "epoch": 5.029103608847497,
+      "grad_norm": 1.2299214601516724,
+      "learning_rate": 0.0002,
+      "loss": 1.0014,
+      "step": 6480
+    },
+    {
+      "epoch": 5.03686457120683,
+      "grad_norm": 1.3125724792480469,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 6490
+    },
+    {
+      "epoch": 5.044625533566162,
+      "grad_norm": 1.1252319812774658,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 6500
+    },
+    {
+      "epoch": 5.052386495925495,
+      "grad_norm": 0.9970866441726685,
+      "learning_rate": 0.0002,
+      "loss": 0.976,
+      "step": 6510
+    },
+    {
+      "epoch": 5.060147458284828,
+      "grad_norm": 1.229069709777832,
+      "learning_rate": 0.0002,
+      "loss": 0.9731,
+      "step": 6520
+    },
+    {
+      "epoch": 5.06790842064416,
+      "grad_norm": 1.2430938482284546,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 6530
+    },
+    {
+      "epoch": 5.0756693830034925,
+      "grad_norm": 1.0522737503051758,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 6540
+    },
+    {
+      "epoch": 5.083430345362825,
+      "grad_norm": 1.108890175819397,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 6550
+    },
+    {
+      "epoch": 5.091191307722157,
+      "grad_norm": 1.156912922859192,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 6560
+    },
+    {
+      "epoch": 5.09895227008149,
+      "grad_norm": 1.405895709991455,
+      "learning_rate": 0.0002,
+      "loss": 1.0415,
+      "step": 6570
+    },
+    {
+      "epoch": 5.106713232440823,
+      "grad_norm": 1.2005155086517334,
+      "learning_rate": 0.0002,
+      "loss": 0.9811,
+      "step": 6580
+    },
+    {
+      "epoch": 5.114474194800155,
+      "grad_norm": 1.181443452835083,
+      "learning_rate": 0.0002,
+      "loss": 0.9862,
+      "step": 6590
+    },
+    {
+      "epoch": 5.122235157159488,
+      "grad_norm": 2.3444771766662598,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 6600
+    },
+    {
+      "epoch": 5.12999611951882,
+      "grad_norm": 1.216988444328308,
+      "learning_rate": 0.0002,
+      "loss": 1.0455,
+      "step": 6610
+    },
+    {
+      "epoch": 5.137757081878153,
+      "grad_norm": 1.369553565979004,
+      "learning_rate": 0.0002,
+      "loss": 1.0549,
+      "step": 6620
+    },
+    {
+      "epoch": 5.145518044237486,
+      "grad_norm": 1.177964687347412,
+      "learning_rate": 0.0002,
+      "loss": 1.0056,
+      "step": 6630
+    },
+    {
+      "epoch": 5.153279006596818,
+      "grad_norm": 1.1397041082382202,
+      "learning_rate": 0.0002,
+      "loss": 1.1025,
+      "step": 6640
+    },
+    {
+      "epoch": 5.161039968956151,
+      "grad_norm": 1.3976861238479614,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 6650
+    },
+    {
+      "epoch": 5.1688009313154835,
+      "grad_norm": 1.4824495315551758,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 6660
+    },
+    {
+      "epoch": 5.1765618936748155,
+      "grad_norm": 1.2653018236160278,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 6670
+    },
+    {
+      "epoch": 5.184322856034148,
+      "grad_norm": 1.3106069564819336,
+      "learning_rate": 0.0002,
+      "loss": 0.9971,
+      "step": 6680
+    },
+    {
+      "epoch": 5.192083818393481,
+      "grad_norm": 1.3140279054641724,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 6690
+    },
+    {
+      "epoch": 5.199844780752813,
+      "grad_norm": 1.3900256156921387,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 6700
+    },
+    {
+      "epoch": 5.207605743112146,
+      "grad_norm": 1.3191124200820923,
+      "learning_rate": 0.0002,
+      "loss": 1.0285,
+      "step": 6710
+    },
+    {
+      "epoch": 5.215366705471478,
+      "grad_norm": 1.176107406616211,
+      "learning_rate": 0.0002,
+      "loss": 0.9921,
+      "step": 6720
+    },
+    {
+      "epoch": 5.223127667830811,
+      "grad_norm": 1.2364883422851562,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 6730
+    },
+    {
+      "epoch": 5.230888630190144,
+      "grad_norm": 1.343022108078003,
+      "learning_rate": 0.0002,
+      "loss": 0.9599,
+      "step": 6740
+    },
+    {
+      "epoch": 5.238649592549476,
+      "grad_norm": 1.2826898097991943,
+      "learning_rate": 0.0002,
+      "loss": 1.0342,
+      "step": 6750
+    },
+    {
+      "epoch": 5.246410554908809,
+      "grad_norm": 1.500257134437561,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 6760
+    },
+    {
+      "epoch": 5.254171517268142,
+      "grad_norm": 1.2605743408203125,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 6770
+    },
+    {
+      "epoch": 5.261932479627474,
+      "grad_norm": 1.2355525493621826,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6780
+    },
+    {
+      "epoch": 5.2696934419868064,
+      "grad_norm": 1.2845789194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.0436,
+      "step": 6790
+    },
+    {
+      "epoch": 5.277454404346139,
+      "grad_norm": 1.3696625232696533,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 6800
+    },
+    {
+      "epoch": 5.285215366705471,
+      "grad_norm": 1.4051260948181152,
+      "learning_rate": 0.0002,
+      "loss": 1.0991,
+      "step": 6810
+    },
+    {
+      "epoch": 5.292976329064804,
+      "grad_norm": 1.266725778579712,
+      "learning_rate": 0.0002,
+      "loss": 1.0987,
+      "step": 6820
+    },
+    {
+      "epoch": 5.300737291424136,
+      "grad_norm": 1.3475236892700195,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 6830
+    },
+    {
+      "epoch": 5.308498253783469,
+      "grad_norm": 1.54409921169281,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 6840
+    },
+    {
+      "epoch": 5.316259216142802,
+      "grad_norm": 1.2391985654830933,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 6850
+    },
+    {
+      "epoch": 5.324020178502134,
+      "grad_norm": 1.2435699701309204,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 6860
+    },
+    {
+      "epoch": 5.331781140861467,
+      "grad_norm": 1.8803037405014038,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6870
+    },
+    {
+      "epoch": 5.3395421032208,
+      "grad_norm": 1.4195542335510254,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 6880
+    },
+    {
+      "epoch": 5.347303065580132,
+      "grad_norm": 1.1853394508361816,
+      "learning_rate": 0.0002,
+      "loss": 1.0273,
+      "step": 6890
+    },
+    {
+      "epoch": 5.355064027939465,
+      "grad_norm": 1.4016530513763428,
+      "learning_rate": 0.0002,
+      "loss": 1.0668,
+      "step": 6900
+    },
+    {
+      "epoch": 5.3628249902987974,
+      "grad_norm": 1.294339895248413,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 6910
+    },
+    {
+      "epoch": 5.370585952658129,
+      "grad_norm": 1.2952708005905151,
+      "learning_rate": 0.0002,
+      "loss": 1.0724,
+      "step": 6920
+    },
+    {
+      "epoch": 5.378346915017462,
+      "grad_norm": 1.1361510753631592,
+      "learning_rate": 0.0002,
+      "loss": 1.0098,
+      "step": 6930
+    },
+    {
+      "epoch": 5.386107877376794,
+      "grad_norm": 1.125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 6940
+    },
+    {
+      "epoch": 5.393868839736127,
+      "grad_norm": 1.1453300714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 6950
+    },
+    {
+      "epoch": 5.40162980209546,
+      "grad_norm": 1.4542768001556396,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6960
+    },
+    {
+      "epoch": 5.409390764454792,
+      "grad_norm": 1.2360988855361938,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6970
+    },
+    {
+      "epoch": 5.417151726814125,
+      "grad_norm": 1.2182754278182983,
+      "learning_rate": 0.0002,
+      "loss": 1.0631,
+      "step": 6980
+    },
+    {
+      "epoch": 5.424912689173458,
+      "grad_norm": 1.2018693685531616,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 6990
+    },
+    {
+      "epoch": 5.43267365153279,
+      "grad_norm": 1.346124291419983,
+      "learning_rate": 0.0002,
+      "loss": 1.108,
+      "step": 7000
+    },
+    {
+      "epoch": 5.440434613892123,
+      "grad_norm": 1.2534189224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7010
+    },
+    {
+      "epoch": 5.448195576251456,
+      "grad_norm": 1.2033339738845825,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 7020
+    },
+    {
+      "epoch": 5.4559565386107876,
+      "grad_norm": 1.2788134813308716,
+      "learning_rate": 0.0002,
+      "loss": 1.0714,
+      "step": 7030
+    },
+    {
+      "epoch": 5.46371750097012,
+      "grad_norm": 1.2751542329788208,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 7040
+    },
+    {
+      "epoch": 5.471478463329452,
+      "grad_norm": 1.3237019777297974,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7050
+    },
+    {
+      "epoch": 5.479239425688785,
+      "grad_norm": 1.4932852983474731,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 7060
+    },
+    {
+      "epoch": 5.487000388048118,
+      "grad_norm": 1.4003876447677612,
+      "learning_rate": 0.0002,
+      "loss": 1.0197,
+      "step": 7070
+    },
+    {
+      "epoch": 5.49476135040745,
+      "grad_norm": 1.404799461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7080
+    },
+    {
+      "epoch": 5.502522312766783,
+      "grad_norm": 1.4486982822418213,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 7090
+    },
+    {
+      "epoch": 5.510283275126116,
+      "grad_norm": 1.1713480949401855,
+      "learning_rate": 0.0002,
+      "loss": 1.0645,
+      "step": 7100
+    },
+    {
+      "epoch": 5.518044237485448,
+      "grad_norm": 1.4062601327896118,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 7110
+    },
+    {
+      "epoch": 5.525805199844781,
+      "grad_norm": 1.211629867553711,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 7120
+    },
+    {
+      "epoch": 5.533566162204114,
+      "grad_norm": 1.2523176670074463,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 7130
+    },
+    {
+      "epoch": 5.541327124563446,
+      "grad_norm": 1.4467198848724365,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 7140
+    },
+    {
+      "epoch": 5.5490880869227786,
+      "grad_norm": 1.5961614847183228,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 7150
+    },
+    {
+      "epoch": 5.5568490492821105,
+      "grad_norm": 1.320656418800354,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 7160
+    },
+    {
+      "epoch": 5.564610011641443,
+      "grad_norm": 1.2423332929611206,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 7170
+    },
+    {
+      "epoch": 5.572370974000776,
+      "grad_norm": 1.2919669151306152,
+      "learning_rate": 0.0002,
+      "loss": 1.0046,
+      "step": 7180
+    },
+    {
+      "epoch": 5.580131936360108,
+      "grad_norm": 1.1678385734558105,
+      "learning_rate": 0.0002,
+      "loss": 1.046,
+      "step": 7190
+    },
+    {
+      "epoch": 5.587892898719441,
+      "grad_norm": 1.4250764846801758,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 7200
+    },
+    {
+      "epoch": 5.595653861078774,
+      "grad_norm": 1.5308716297149658,
+      "learning_rate": 0.0002,
+      "loss": 1.1254,
+      "step": 7210
+    },
+    {
+      "epoch": 5.603414823438106,
+      "grad_norm": 1.2678815126419067,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7220
+    },
+    {
+      "epoch": 5.611175785797439,
+      "grad_norm": 1.127856969833374,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 7230
+    },
+    {
+      "epoch": 5.618936748156772,
+      "grad_norm": 1.3832560777664185,
+      "learning_rate": 0.0002,
+      "loss": 1.0647,
+      "step": 7240
+    },
+    {
+      "epoch": 5.626697710516104,
+      "grad_norm": 1.3226919174194336,
+      "learning_rate": 0.0002,
+      "loss": 1.0658,
+      "step": 7250
+    },
+    {
+      "epoch": 5.634458672875437,
+      "grad_norm": 1.3418006896972656,
+      "learning_rate": 0.0002,
+      "loss": 1.1175,
+      "step": 7260
+    },
+    {
+      "epoch": 5.642219635234769,
+      "grad_norm": 1.2625300884246826,
+      "learning_rate": 0.0002,
+      "loss": 1.0956,
+      "step": 7270
+    },
+    {
+      "epoch": 5.6499805975941015,
+      "grad_norm": 1.1579464673995972,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 7280
+    },
+    {
+      "epoch": 5.657741559953434,
+      "grad_norm": 1.4998650550842285,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 7290
+    },
+    {
+      "epoch": 5.665502522312766,
+      "grad_norm": 1.2670758962631226,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7300
+    },
+    {
+      "epoch": 5.673263484672099,
+      "grad_norm": 1.2959760427474976,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 7310
+    },
+    {
+      "epoch": 5.681024447031432,
+      "grad_norm": 1.2460671663284302,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 7320
+    },
+    {
+      "epoch": 5.688785409390764,
+      "grad_norm": 1.1313989162445068,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 7330
+    },
+    {
+      "epoch": 5.696546371750097,
+      "grad_norm": 1.282527208328247,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 7340
+    },
+    {
+      "epoch": 5.70430733410943,
+      "grad_norm": 1.3380206823349,
+      "learning_rate": 0.0002,
+      "loss": 1.1315,
+      "step": 7350
+    },
+    {
+      "epoch": 5.712068296468762,
+      "grad_norm": 1.1648279428482056,
+      "learning_rate": 0.0002,
+      "loss": 1.0949,
+      "step": 7360
+    },
+    {
+      "epoch": 5.719829258828095,
+      "grad_norm": 1.3059816360473633,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 7370
+    },
+    {
+      "epoch": 5.727590221187427,
+      "grad_norm": 1.1905046701431274,
+      "learning_rate": 0.0002,
+      "loss": 1.1496,
+      "step": 7380
+    },
+    {
+      "epoch": 5.73535118354676,
+      "grad_norm": 1.4089630842208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1356,
+      "step": 7390
+    },
+    {
+      "epoch": 5.7431121459060925,
+      "grad_norm": 1.256721019744873,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 7400
+    },
+    {
+      "epoch": 5.7508731082654245,
+      "grad_norm": 1.1915162801742554,
+      "learning_rate": 0.0002,
+      "loss": 1.0682,
+      "step": 7410
+    },
+    {
+      "epoch": 5.758634070624757,
+      "grad_norm": 1.1935480833053589,
+      "learning_rate": 0.0002,
+      "loss": 1.1257,
+      "step": 7420
+    },
+    {
+      "epoch": 5.76639503298409,
+      "grad_norm": 1.1761008501052856,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 7430
+    },
+    {
+      "epoch": 5.774155995343422,
+      "grad_norm": 1.2540549039840698,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 7440
+    },
+    {
+      "epoch": 5.781916957702755,
+      "grad_norm": 1.5295120477676392,
+      "learning_rate": 0.0002,
+      "loss": 1.1527,
+      "step": 7450
+    },
+    {
+      "epoch": 5.789677920062088,
+      "grad_norm": 1.1081160306930542,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 7460
+    },
+    {
+      "epoch": 5.79743888242142,
+      "grad_norm": 1.4381253719329834,
+      "learning_rate": 0.0002,
+      "loss": 1.1304,
+      "step": 7470
+    },
+    {
+      "epoch": 5.805199844780753,
+      "grad_norm": 1.3079341650009155,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 7480
+    },
+    {
+      "epoch": 5.812960807140085,
+      "grad_norm": 1.1372792720794678,
+      "learning_rate": 0.0002,
+      "loss": 1.0544,
+      "step": 7490
+    },
+    {
+      "epoch": 5.820721769499418,
+      "grad_norm": 1.3221744298934937,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 7500
+    },
+    {
+      "epoch": 5.828482731858751,
+      "grad_norm": 1.3436939716339111,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 7510
+    },
+    {
+      "epoch": 5.8362436942180835,
+      "grad_norm": 1.3916879892349243,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 7520
+    },
+    {
+      "epoch": 5.8440046565774155,
+      "grad_norm": 1.2463704347610474,
+      "learning_rate": 0.0002,
+      "loss": 1.0816,
+      "step": 7530
+    },
+    {
+      "epoch": 5.851765618936748,
+      "grad_norm": 1.097051739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.0745,
+      "step": 7540
+    },
+    {
+      "epoch": 5.85952658129608,
+      "grad_norm": 1.1554739475250244,
+      "learning_rate": 0.0002,
+      "loss": 1.1454,
+      "step": 7550
+    },
+    {
+      "epoch": 5.867287543655413,
+      "grad_norm": 1.2384694814682007,
+      "learning_rate": 0.0002,
+      "loss": 1.0953,
+      "step": 7560
+    },
+    {
+      "epoch": 5.875048506014746,
+      "grad_norm": 1.142815351486206,
+      "learning_rate": 0.0002,
+      "loss": 1.1734,
+      "step": 7570
+    },
+    {
+      "epoch": 5.882809468374078,
+      "grad_norm": 1.3637062311172485,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 7580
+    },
+    {
+      "epoch": 5.890570430733411,
+      "grad_norm": 1.2449073791503906,
+      "learning_rate": 0.0002,
+      "loss": 1.0781,
+      "step": 7590
+    },
+    {
+      "epoch": 5.898331393092743,
+      "grad_norm": 1.358058214187622,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 7600
+    },
+    {
+      "epoch": 5.906092355452076,
+      "grad_norm": 1.264655351638794,
+      "learning_rate": 0.0002,
+      "loss": 1.0779,
+      "step": 7610
+    },
+    {
+      "epoch": 5.913853317811409,
+      "grad_norm": 1.3186019659042358,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 7620
+    },
+    {
+      "epoch": 5.921614280170742,
+      "grad_norm": 1.4111460447311401,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 7630
+    },
+    {
+      "epoch": 5.929375242530074,
+      "grad_norm": 1.1078972816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.1765,
+      "step": 7640
+    },
+    {
+      "epoch": 5.9371362048894065,
+      "grad_norm": 1.2742213010787964,
+      "learning_rate": 0.0002,
+      "loss": 1.1305,
+      "step": 7650
+    },
+    {
+      "epoch": 5.9448971672487385,
+      "grad_norm": 1.3412781953811646,
+      "learning_rate": 0.0002,
+      "loss": 1.144,
+      "step": 7660
+    },
+    {
+      "epoch": 5.952658129608071,
+      "grad_norm": 1.123005986213684,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 7670
+    },
+    {
+      "epoch": 5.960419091967404,
+      "grad_norm": 1.2203444242477417,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 7680
+    },
+    {
+      "epoch": 5.968180054326736,
+      "grad_norm": 1.341011643409729,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 7690
+    },
+    {
+      "epoch": 5.975941016686069,
+      "grad_norm": 1.2689454555511475,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 7700
+    },
+    {
+      "epoch": 5.983701979045401,
+      "grad_norm": 1.1518112421035767,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 7710
+    },
+    {
+      "epoch": 5.991462941404734,
+      "grad_norm": 1.3698320388793945,
+      "learning_rate": 0.0002,
+      "loss": 1.1868,
+      "step": 7720
+    },
+    {
+      "epoch": 5.999223903764067,
+      "grad_norm": 1.2812788486480713,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 7730
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.252762794494629,
+      "eval_runtime": 114.8471,
+      "eval_samples_per_second": 4.415,
+      "eval_steps_per_second": 0.557,
+      "step": 7731
+    },
+    {
+      "epoch": 6.006984866123399,
+      "grad_norm": 1.8642009496688843,
+      "learning_rate": 0.0002,
+      "loss": 0.8629,
+      "step": 7740
+    },
+    {
+      "epoch": 6.014745828482732,
+      "grad_norm": 1.7081232070922852,
+      "learning_rate": 0.0002,
+      "loss": 0.8435,
+      "step": 7750
+    },
+    {
+      "epoch": 6.022506790842065,
+      "grad_norm": 1.6233899593353271,
+      "learning_rate": 0.0002,
+      "loss": 0.7729,
+      "step": 7760
+    },
+    {
+      "epoch": 6.030267753201397,
+      "grad_norm": 1.5111888647079468,
+      "learning_rate": 0.0002,
+      "loss": 0.7907,
+      "step": 7770
+    },
+    {
+      "epoch": 6.0380287155607295,
+      "grad_norm": 1.5278418064117432,
+      "learning_rate": 0.0002,
+      "loss": 0.7908,
+      "step": 7780
+    },
+    {
+      "epoch": 6.045789677920062,
+      "grad_norm": 1.5932185649871826,
+      "learning_rate": 0.0002,
+      "loss": 0.835,
+      "step": 7790
+    },
+    {
+      "epoch": 6.053550640279394,
+      "grad_norm": 1.5990597009658813,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 7800
+    },
+    {
+      "epoch": 6.061311602638727,
+      "grad_norm": 1.7498669624328613,
+      "learning_rate": 0.0002,
+      "loss": 0.8559,
+      "step": 7810
+    },
+    {
+      "epoch": 6.06907256499806,
+      "grad_norm": 1.6105555295944214,
+      "learning_rate": 0.0002,
+      "loss": 0.8069,
+      "step": 7820
+    },
+    {
+      "epoch": 6.076833527357392,
+      "grad_norm": 1.5214293003082275,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 7830
+    },
+    {
+      "epoch": 6.084594489716725,
+      "grad_norm": 1.6586973667144775,
+      "learning_rate": 0.0002,
+      "loss": 0.8328,
+      "step": 7840
+    },
+    {
+      "epoch": 6.092355452076057,
+      "grad_norm": 1.467391848564148,
+      "learning_rate": 0.0002,
+      "loss": 0.8415,
+      "step": 7850
+    },
+    {
+      "epoch": 6.10011641443539,
+      "grad_norm": 1.537361741065979,
+      "learning_rate": 0.0002,
+      "loss": 0.8274,
+      "step": 7860
+    },
+    {
+      "epoch": 6.107877376794723,
+      "grad_norm": 1.621764898300171,
+      "learning_rate": 0.0002,
+      "loss": 0.8011,
+      "step": 7870
+    },
+    {
+      "epoch": 6.115638339154055,
+      "grad_norm": 1.583751916885376,
+      "learning_rate": 0.0002,
+      "loss": 0.8556,
+      "step": 7880
+    },
+    {
+      "epoch": 6.123399301513388,
+      "grad_norm": 1.6199619770050049,
+      "learning_rate": 0.0002,
+      "loss": 0.8829,
+      "step": 7890
+    },
+    {
+      "epoch": 6.1311602638727205,
+      "grad_norm": 1.6163095235824585,
+      "learning_rate": 0.0002,
+      "loss": 0.8226,
+      "step": 7900
+    },
+    {
+      "epoch": 6.1389212262320525,
+      "grad_norm": 1.6120976209640503,
+      "learning_rate": 0.0002,
+      "loss": 0.8203,
+      "step": 7910
+    },
+    {
+      "epoch": 6.146682188591385,
+      "grad_norm": 1.7886850833892822,
+      "learning_rate": 0.0002,
+      "loss": 0.7915,
+      "step": 7920
+    },
+    {
+      "epoch": 6.154443150950718,
+      "grad_norm": 1.408303141593933,
+      "learning_rate": 0.0002,
+      "loss": 0.7808,
+      "step": 7930
+    },
+    {
+      "epoch": 6.16220411331005,
+      "grad_norm": 1.6048113107681274,
+      "learning_rate": 0.0002,
+      "loss": 0.8404,
+      "step": 7940
+    },
+    {
+      "epoch": 6.169965075669383,
+      "grad_norm": 1.424306869506836,
+      "learning_rate": 0.0002,
+      "loss": 0.8705,
+      "step": 7950
+    },
+    {
+      "epoch": 6.177726038028716,
+      "grad_norm": 1.4453672170639038,
+      "learning_rate": 0.0002,
+      "loss": 0.8177,
+      "step": 7960
+    },
+    {
+      "epoch": 6.185487000388048,
+      "grad_norm": 1.3157061338424683,
+      "learning_rate": 0.0002,
+      "loss": 0.8182,
+      "step": 7970
+    },
+    {
+      "epoch": 6.193247962747381,
+      "grad_norm": 1.330541729927063,
+      "learning_rate": 0.0002,
+      "loss": 0.891,
+      "step": 7980
+    },
+    {
+      "epoch": 6.201008925106713,
+      "grad_norm": 1.6306229829788208,
+      "learning_rate": 0.0002,
+      "loss": 0.8599,
+      "step": 7990
+    },
+    {
+      "epoch": 6.208769887466046,
+      "grad_norm": 1.6332136392593384,
+      "learning_rate": 0.0002,
+      "loss": 0.9069,
+      "step": 8000
+    },
+    {
+      "epoch": 6.216530849825379,
+      "grad_norm": 1.708613395690918,
+      "learning_rate": 0.0002,
+      "loss": 0.83,
+      "step": 8010
+    },
+    {
+      "epoch": 6.224291812184711,
+      "grad_norm": 1.6637346744537354,
+      "learning_rate": 0.0002,
+      "loss": 0.8509,
+      "step": 8020
+    },
+    {
+      "epoch": 6.2320527745440435,
+      "grad_norm": 1.5675315856933594,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 8030
+    },
+    {
+      "epoch": 6.239813736903376,
+      "grad_norm": 1.5826327800750732,
+      "learning_rate": 0.0002,
+      "loss": 0.8491,
+      "step": 8040
+    },
+    {
+      "epoch": 6.247574699262708,
+      "grad_norm": 1.7382984161376953,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 8050
+    },
+    {
+      "epoch": 6.255335661622041,
+      "grad_norm": 1.5272295475006104,
+      "learning_rate": 0.0002,
+      "loss": 0.8795,
+      "step": 8060
+    },
+    {
+      "epoch": 6.263096623981374,
+      "grad_norm": 1.8195022344589233,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 8070
+    },
+    {
+      "epoch": 6.270857586340706,
+      "grad_norm": 1.679901361465454,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 8080
+    },
+    {
+      "epoch": 6.278618548700039,
+      "grad_norm": 1.4921348094940186,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 8090
+    },
+    {
+      "epoch": 6.286379511059371,
+      "grad_norm": 1.4627857208251953,
+      "learning_rate": 0.0002,
+      "loss": 0.899,
+      "step": 8100
+    },
+    {
+      "epoch": 6.294140473418704,
+      "grad_norm": 1.3528631925582886,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 8110
+    },
+    {
+      "epoch": 6.301901435778037,
+      "grad_norm": 1.6863102912902832,
+      "learning_rate": 0.0002,
+      "loss": 0.9355,
+      "step": 8120
+    },
+    {
+      "epoch": 6.309662398137369,
+      "grad_norm": 1.6178052425384521,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 8130
+    },
+    {
+      "epoch": 6.317423360496702,
+      "grad_norm": 1.7626280784606934,
+      "learning_rate": 0.0002,
+      "loss": 0.9182,
+      "step": 8140
+    },
+    {
+      "epoch": 6.3251843228560345,
+      "grad_norm": 1.7188845872879028,
+      "learning_rate": 0.0002,
+      "loss": 0.8886,
+      "step": 8150
+    },
+    {
+      "epoch": 6.3329452852153665,
+      "grad_norm": 1.5777133703231812,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 8160
+    },
+    {
+      "epoch": 6.340706247574699,
+      "grad_norm": 1.7653207778930664,
+      "learning_rate": 0.0002,
+      "loss": 0.9247,
+      "step": 8170
+    },
+    {
+      "epoch": 6.348467209934032,
+      "grad_norm": 1.6861237287521362,
+      "learning_rate": 0.0002,
+      "loss": 0.8003,
+      "step": 8180
+    },
+    {
+      "epoch": 6.356228172293364,
+      "grad_norm": 1.6318124532699585,
+      "learning_rate": 0.0002,
+      "loss": 0.884,
+      "step": 8190
+    },
+    {
+      "epoch": 6.363989134652697,
+      "grad_norm": 1.6192939281463623,
+      "learning_rate": 0.0002,
+      "loss": 0.8341,
+      "step": 8200
+    },
+    {
+      "epoch": 6.371750097012029,
+      "grad_norm": 1.7641773223876953,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 8210
+    },
+    {
+      "epoch": 6.379511059371362,
+      "grad_norm": 1.6470493078231812,
+      "learning_rate": 0.0002,
+      "loss": 0.8582,
+      "step": 8220
+    },
+    {
+      "epoch": 6.387272021730695,
+      "grad_norm": 1.5898468494415283,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 8230
+    },
+    {
+      "epoch": 6.395032984090027,
+      "grad_norm": 1.8025981187820435,
+      "learning_rate": 0.0002,
+      "loss": 0.9658,
+      "step": 8240
+    },
+    {
+      "epoch": 6.40279394644936,
+      "grad_norm": 1.7035106420516968,
+      "learning_rate": 0.0002,
+      "loss": 0.8953,
+      "step": 8250
+    },
+    {
+      "epoch": 6.410554908808693,
+      "grad_norm": 1.5968799591064453,
+      "learning_rate": 0.0002,
+      "loss": 0.9193,
+      "step": 8260
+    },
+    {
+      "epoch": 6.418315871168025,
+      "grad_norm": 1.7492800951004028,
+      "learning_rate": 0.0002,
+      "loss": 0.929,
+      "step": 8270
+    },
+    {
+      "epoch": 6.4260768335273575,
+      "grad_norm": 1.6914138793945312,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 8280
+    },
+    {
+      "epoch": 6.43383779588669,
+      "grad_norm": 1.5761380195617676,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 8290
+    },
+    {
+      "epoch": 6.441598758246022,
+      "grad_norm": 1.5164411067962646,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 8300
+    },
+    {
+      "epoch": 6.449359720605355,
+      "grad_norm": 1.6600215435028076,
+      "learning_rate": 0.0002,
+      "loss": 0.88,
+      "step": 8310
+    },
+    {
+      "epoch": 6.457120682964687,
+      "grad_norm": 1.2477679252624512,
+      "learning_rate": 0.0002,
+      "loss": 0.9113,
+      "step": 8320
+    },
+    {
+      "epoch": 6.46488164532402,
+      "grad_norm": 1.3698599338531494,
+      "learning_rate": 0.0002,
+      "loss": 0.8822,
+      "step": 8330
+    },
+    {
+      "epoch": 6.472642607683353,
+      "grad_norm": 1.4847341775894165,
+      "learning_rate": 0.0002,
+      "loss": 0.9295,
+      "step": 8340
+    },
+    {
+      "epoch": 6.480403570042685,
+      "grad_norm": 1.4713412523269653,
+      "learning_rate": 0.0002,
+      "loss": 0.9243,
+      "step": 8350
+    },
+    {
+      "epoch": 6.488164532402018,
+      "grad_norm": 1.334523320198059,
+      "learning_rate": 0.0002,
+      "loss": 0.9102,
+      "step": 8360
+    },
+    {
+      "epoch": 6.495925494761351,
+      "grad_norm": 2.0054359436035156,
+      "learning_rate": 0.0002,
+      "loss": 0.8563,
+      "step": 8370
+    },
+    {
+      "epoch": 6.503686457120683,
+      "grad_norm": 1.560014247894287,
+      "learning_rate": 0.0002,
+      "loss": 0.9759,
+      "step": 8380
+    },
+    {
+      "epoch": 6.511447419480016,
+      "grad_norm": 1.518526554107666,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 8390
+    },
+    {
+      "epoch": 6.5192083818393485,
+      "grad_norm": 1.3841272592544556,
+      "learning_rate": 0.0002,
+      "loss": 0.937,
+      "step": 8400
+    },
+    {
+      "epoch": 6.5269693441986805,
+      "grad_norm": 1.5191527605056763,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 8410
+    },
+    {
+      "epoch": 6.534730306558013,
+      "grad_norm": 1.5275579690933228,
+      "learning_rate": 0.0002,
+      "loss": 0.8899,
+      "step": 8420
+    },
+    {
+      "epoch": 6.542491268917345,
+      "grad_norm": 1.621590256690979,
+      "learning_rate": 0.0002,
+      "loss": 0.9291,
+      "step": 8430
+    },
+    {
+      "epoch": 6.550252231276678,
+      "grad_norm": 1.7939082384109497,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 8440
+    },
+    {
+      "epoch": 6.558013193636011,
+      "grad_norm": 1.4542964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8896,
+      "step": 8450
+    },
+    {
+      "epoch": 6.565774155995343,
+      "grad_norm": 1.5458455085754395,
+      "learning_rate": 0.0002,
+      "loss": 0.9393,
+      "step": 8460
+    },
+    {
+      "epoch": 6.573535118354676,
+      "grad_norm": 1.550359845161438,
+      "learning_rate": 0.0002,
+      "loss": 0.9028,
+      "step": 8470
+    },
+    {
+      "epoch": 6.581296080714009,
+      "grad_norm": 1.527757167816162,
+      "learning_rate": 0.0002,
+      "loss": 0.9271,
+      "step": 8480
+    },
+    {
+      "epoch": 6.589057043073341,
+      "grad_norm": 1.4683486223220825,
+      "learning_rate": 0.0002,
+      "loss": 0.966,
+      "step": 8490
+    },
+    {
+      "epoch": 6.596818005432674,
+      "grad_norm": 1.5057084560394287,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 8500
+    },
+    {
+      "epoch": 6.604578967792007,
+      "grad_norm": 1.648289442062378,
+      "learning_rate": 0.0002,
+      "loss": 0.9235,
+      "step": 8510
+    },
+    {
+      "epoch": 6.612339930151339,
+      "grad_norm": 1.578914761543274,
+      "learning_rate": 0.0002,
+      "loss": 0.9113,
+      "step": 8520
+    },
+    {
+      "epoch": 6.6201008925106715,
+      "grad_norm": 1.5064080953598022,
+      "learning_rate": 0.0002,
+      "loss": 0.8894,
+      "step": 8530
+    },
+    {
+      "epoch": 6.6278618548700035,
+      "grad_norm": 1.5717744827270508,
+      "learning_rate": 0.0002,
+      "loss": 0.8981,
+      "step": 8540
+    },
+    {
+      "epoch": 6.635622817229336,
+      "grad_norm": 1.7954767942428589,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 8550
+    },
+    {
+      "epoch": 6.643383779588669,
+      "grad_norm": 1.6172343492507935,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 8560
+    },
+    {
+      "epoch": 6.651144741948001,
+      "grad_norm": 1.6627886295318604,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 8570
+    },
+    {
+      "epoch": 6.658905704307334,
+      "grad_norm": 1.5264919996261597,
+      "learning_rate": 0.0002,
+      "loss": 0.959,
+      "step": 8580
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.609248161315918,
+      "learning_rate": 0.0002,
+      "loss": 0.9103,
+      "step": 8590
+    },
+    {
+      "epoch": 6.674427629025999,
+      "grad_norm": 1.5474581718444824,
+      "learning_rate": 0.0002,
+      "loss": 0.9395,
+      "step": 8600
+    },
+    {
+      "epoch": 6.682188591385332,
+      "grad_norm": 1.6294898986816406,
+      "learning_rate": 0.0002,
+      "loss": 0.9018,
+      "step": 8610
+    },
+    {
+      "epoch": 6.689949553744665,
+      "grad_norm": 1.612615942955017,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 8620
+    },
+    {
+      "epoch": 6.697710516103997,
+      "grad_norm": 1.741325855255127,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 8630
+    },
+    {
+      "epoch": 6.70547147846333,
+      "grad_norm": 1.5089004039764404,
+      "learning_rate": 0.0002,
+      "loss": 1.0475,
+      "step": 8640
+    },
+    {
+      "epoch": 6.713232440822662,
+      "grad_norm": 1.4725582599639893,
+      "learning_rate": 0.0002,
+      "loss": 1.0009,
+      "step": 8650
+    },
+    {
+      "epoch": 6.7209934031819945,
+      "grad_norm": 1.6992095708847046,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 8660
+    },
+    {
+      "epoch": 6.728754365541327,
+      "grad_norm": 1.5938470363616943,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 8670
+    },
+    {
+      "epoch": 6.736515327900659,
+      "grad_norm": 1.58723783493042,
+      "learning_rate": 0.0002,
+      "loss": 0.9411,
+      "step": 8680
+    },
+    {
+      "epoch": 6.744276290259992,
+      "grad_norm": 1.514389991760254,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 8690
+    },
+    {
+      "epoch": 6.752037252619325,
+      "grad_norm": 1.6799157857894897,
+      "learning_rate": 0.0002,
+      "loss": 0.9283,
+      "step": 8700
+    },
+    {
+      "epoch": 6.759798214978657,
+      "grad_norm": 1.5436359643936157,
+      "learning_rate": 0.0002,
+      "loss": 0.9138,
+      "step": 8710
+    },
+    {
+      "epoch": 6.76755917733799,
+      "grad_norm": 1.477137565612793,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8720
+    },
+    {
+      "epoch": 6.775320139697323,
+      "grad_norm": 1.7383503913879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0044,
+      "step": 8730
+    },
+    {
+      "epoch": 6.783081102056655,
+      "grad_norm": 1.8000324964523315,
+      "learning_rate": 0.0002,
+      "loss": 0.9492,
+      "step": 8740
+    },
+    {
+      "epoch": 6.790842064415988,
+      "grad_norm": 1.3099453449249268,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 8750
+    },
+    {
+      "epoch": 6.79860302677532,
+      "grad_norm": 1.8775172233581543,
+      "learning_rate": 0.0002,
+      "loss": 0.9709,
+      "step": 8760
+    },
+    {
+      "epoch": 6.806363989134653,
+      "grad_norm": 1.5832085609436035,
+      "learning_rate": 0.0002,
+      "loss": 0.9356,
+      "step": 8770
+    },
+    {
+      "epoch": 6.8141249514939854,
+      "grad_norm": 1.4903252124786377,
+      "learning_rate": 0.0002,
+      "loss": 0.9397,
+      "step": 8780
+    },
+    {
+      "epoch": 6.821885913853317,
+      "grad_norm": 1.6360470056533813,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 8790
+    },
+    {
+      "epoch": 6.82964687621265,
+      "grad_norm": 1.5457707643508911,
+      "learning_rate": 0.0002,
+      "loss": 0.957,
+      "step": 8800
+    },
+    {
+      "epoch": 6.837407838571983,
+      "grad_norm": 1.5449066162109375,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 8810
+    },
+    {
+      "epoch": 6.845168800931315,
+      "grad_norm": 1.618337631225586,
+      "learning_rate": 0.0002,
+      "loss": 1.0007,
+      "step": 8820
+    },
+    {
+      "epoch": 6.852929763290648,
+      "grad_norm": 1.38296639919281,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 8830
+    },
+    {
+      "epoch": 6.860690725649981,
+      "grad_norm": 1.6427991390228271,
+      "learning_rate": 0.0002,
+      "loss": 0.9349,
+      "step": 8840
+    },
+    {
+      "epoch": 6.868451688009313,
+      "grad_norm": 1.4980270862579346,
+      "learning_rate": 0.0002,
+      "loss": 1.0194,
+      "step": 8850
+    },
+    {
+      "epoch": 6.876212650368646,
+      "grad_norm": 1.3800020217895508,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 8860
+    },
+    {
+      "epoch": 6.883973612727978,
+      "grad_norm": 1.5971838235855103,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 8870
+    },
+    {
+      "epoch": 6.891734575087311,
+      "grad_norm": 1.4429489374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.0105,
+      "step": 8880
+    },
+    {
+      "epoch": 6.899495537446644,
+      "grad_norm": 1.4959166049957275,
+      "learning_rate": 0.0002,
+      "loss": 0.9143,
+      "step": 8890
+    },
+    {
+      "epoch": 6.907256499805976,
+      "grad_norm": 1.5776222944259644,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 8900
+    },
+    {
+      "epoch": 6.915017462165308,
+      "grad_norm": 1.510412573814392,
+      "learning_rate": 0.0002,
+      "loss": 0.9256,
+      "step": 8910
+    },
+    {
+      "epoch": 6.922778424524641,
+      "grad_norm": 1.7216295003890991,
+      "learning_rate": 0.0002,
+      "loss": 1.0095,
+      "step": 8920
+    },
+    {
+      "epoch": 6.930539386883973,
+      "grad_norm": 1.830762505531311,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 8930
+    },
+    {
+      "epoch": 6.938300349243306,
+      "grad_norm": 1.3472434282302856,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 8940
+    },
+    {
+      "epoch": 6.946061311602639,
+      "grad_norm": 1.5748040676116943,
+      "learning_rate": 0.0002,
+      "loss": 0.9718,
+      "step": 8950
+    },
+    {
+      "epoch": 6.953822273961971,
+      "grad_norm": 1.5317506790161133,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 8960
+    },
+    {
+      "epoch": 6.961583236321304,
+      "grad_norm": 1.5565721988677979,
+      "learning_rate": 0.0002,
+      "loss": 0.9513,
+      "step": 8970
+    },
+    {
+      "epoch": 6.969344198680636,
+      "grad_norm": 1.5288970470428467,
+      "learning_rate": 0.0002,
+      "loss": 0.9118,
+      "step": 8980
+    },
+    {
+      "epoch": 6.977105161039969,
+      "grad_norm": 1.562624454498291,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 8990
+    },
+    {
+      "epoch": 6.984866123399302,
+      "grad_norm": 1.3777633905410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9929,
+      "step": 9000
+    },
+    {
+      "epoch": 6.992627085758635,
+      "grad_norm": 1.5868972539901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 9010
+    },
+    {
+      "epoch": 6.9996119518820334,
+      "eval_loss": 2.4372169971466064,
+      "eval_runtime": 113.8966,
+      "eval_samples_per_second": 4.451,
+      "eval_steps_per_second": 0.562,
+      "step": 9019
+    },
+    {
+      "epoch": 7.0003880481179666,
+      "grad_norm": 1.3035310506820679,
+      "learning_rate": 0.0002,
+      "loss": 0.9911,
+      "step": 9020
+    },
+    {
+      "epoch": 7.008149010477299,
+      "grad_norm": 1.5640218257904053,
+      "learning_rate": 0.0002,
+      "loss": 0.6585,
+      "step": 9030
+    },
+    {
+      "epoch": 7.015909972836631,
+      "grad_norm": 1.9529098272323608,
+      "learning_rate": 0.0002,
+      "loss": 0.6507,
+      "step": 9040
+    },
+    {
+      "epoch": 7.023670935195964,
+      "grad_norm": 1.6257457733154297,
+      "learning_rate": 0.0002,
+      "loss": 0.6335,
+      "step": 9050
+    },
+    {
+      "epoch": 7.031431897555297,
+      "grad_norm": 1.8028602600097656,
+      "learning_rate": 0.0002,
+      "loss": 0.6752,
+      "step": 9060
+    },
+    {
+      "epoch": 7.039192859914629,
+      "grad_norm": 1.4882043600082397,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 9070
+    },
+    {
+      "epoch": 7.046953822273962,
+      "grad_norm": 2.10062837600708,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 9080
+    },
+    {
+      "epoch": 7.054714784633295,
+      "grad_norm": 1.6754050254821777,
+      "learning_rate": 0.0002,
+      "loss": 0.643,
+      "step": 9090
+    },
+    {
+      "epoch": 7.062475746992627,
+      "grad_norm": 1.9425220489501953,
+      "learning_rate": 0.0002,
+      "loss": 0.6461,
+      "step": 9100
+    },
+    {
+      "epoch": 7.07023670935196,
+      "grad_norm": 1.9451842308044434,
+      "learning_rate": 0.0002,
+      "loss": 0.6441,
+      "step": 9110
+    },
+    {
+      "epoch": 7.077997671711292,
+      "grad_norm": 2.203806161880493,
+      "learning_rate": 0.0002,
+      "loss": 0.6872,
+      "step": 9120
+    },
+    {
+      "epoch": 7.085758634070625,
+      "grad_norm": 1.7248806953430176,
+      "learning_rate": 0.0002,
+      "loss": 0.6768,
+      "step": 9130
+    },
+    {
+      "epoch": 7.0935195964299576,
+      "grad_norm": 1.5713436603546143,
+      "learning_rate": 0.0002,
+      "loss": 0.6911,
+      "step": 9140
+    },
+    {
+      "epoch": 7.1012805587892895,
+      "grad_norm": 2.0308637619018555,
+      "learning_rate": 0.0002,
+      "loss": 0.6689,
+      "step": 9150
+    },
+    {
+      "epoch": 7.109041521148622,
+      "grad_norm": 1.7522761821746826,
+      "learning_rate": 0.0002,
+      "loss": 0.6617,
+      "step": 9160
+    },
+    {
+      "epoch": 7.116802483507955,
+      "grad_norm": 1.9685192108154297,
+      "learning_rate": 0.0002,
+      "loss": 0.669,
+      "step": 9170
+    },
+    {
+      "epoch": 7.124563445867287,
+      "grad_norm": 2.069120407104492,
+      "learning_rate": 0.0002,
+      "loss": 0.6727,
+      "step": 9180
+    },
+    {
+      "epoch": 7.13232440822662,
+      "grad_norm": 1.7211129665374756,
+      "learning_rate": 0.0002,
+      "loss": 0.6553,
+      "step": 9190
+    },
+    {
+      "epoch": 7.140085370585953,
+      "grad_norm": 1.7535923719406128,
+      "learning_rate": 0.0002,
+      "loss": 0.6827,
+      "step": 9200
+    },
+    {
+      "epoch": 7.147846332945285,
+      "grad_norm": 1.7181583642959595,
+      "learning_rate": 0.0002,
+      "loss": 0.6698,
+      "step": 9210
+    },
+    {
+      "epoch": 7.155607295304618,
+      "grad_norm": 1.7778624296188354,
+      "learning_rate": 0.0002,
+      "loss": 0.7056,
+      "step": 9220
+    },
+    {
+      "epoch": 7.16336825766395,
+      "grad_norm": 1.8051576614379883,
+      "learning_rate": 0.0002,
+      "loss": 0.7326,
+      "step": 9230
+    },
+    {
+      "epoch": 7.171129220023283,
+      "grad_norm": 1.9704501628875732,
+      "learning_rate": 0.0002,
+      "loss": 0.6858,
+      "step": 9240
+    },
+    {
+      "epoch": 7.178890182382616,
+      "grad_norm": 1.8750483989715576,
+      "learning_rate": 0.0002,
+      "loss": 0.7029,
+      "step": 9250
+    },
+    {
+      "epoch": 7.186651144741948,
+      "grad_norm": 1.790107011795044,
+      "learning_rate": 0.0002,
+      "loss": 0.7073,
+      "step": 9260
+    },
+    {
+      "epoch": 7.1944121071012805,
+      "grad_norm": 1.9681477546691895,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 9270
+    },
+    {
+      "epoch": 7.202173069460613,
+      "grad_norm": 1.7811331748962402,
+      "learning_rate": 0.0002,
+      "loss": 0.7074,
+      "step": 9280
+    },
+    {
+      "epoch": 7.209934031819945,
+      "grad_norm": 1.9861894845962524,
+      "learning_rate": 0.0002,
+      "loss": 0.6959,
+      "step": 9290
+    },
+    {
+      "epoch": 7.217694994179278,
+      "grad_norm": 1.8615444898605347,
+      "learning_rate": 0.0002,
+      "loss": 0.6463,
+      "step": 9300
+    },
+    {
+      "epoch": 7.225455956538611,
+      "grad_norm": 1.899372935295105,
+      "learning_rate": 0.0002,
+      "loss": 0.7071,
+      "step": 9310
+    },
+    {
+      "epoch": 7.233216918897943,
+      "grad_norm": 1.8131160736083984,
+      "learning_rate": 0.0002,
+      "loss": 0.6883,
+      "step": 9320
+    },
+    {
+      "epoch": 7.240977881257276,
+      "grad_norm": 1.5020049810409546,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 9330
+    },
+    {
+      "epoch": 7.248738843616608,
+      "grad_norm": 1.7210577726364136,
+      "learning_rate": 0.0002,
+      "loss": 0.7197,
+      "step": 9340
+    },
+    {
+      "epoch": 7.256499805975941,
+      "grad_norm": 2.003021001815796,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 9350
+    },
+    {
+      "epoch": 7.264260768335274,
+      "grad_norm": 1.6632959842681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6746,
+      "step": 9360
+    },
+    {
+      "epoch": 7.272021730694606,
+      "grad_norm": 1.7717185020446777,
+      "learning_rate": 0.0002,
+      "loss": 0.7072,
+      "step": 9370
+    },
+    {
+      "epoch": 7.279782693053939,
+      "grad_norm": 1.8554900884628296,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 9380
+    },
+    {
+      "epoch": 7.2875436554132715,
+      "grad_norm": 1.889708399772644,
+      "learning_rate": 0.0002,
+      "loss": 0.6983,
+      "step": 9390
+    },
+    {
+      "epoch": 7.2953046177726035,
+      "grad_norm": 1.8426263332366943,
+      "learning_rate": 0.0002,
+      "loss": 0.7495,
+      "step": 9400
+    },
+    {
+      "epoch": 7.303065580131936,
+      "grad_norm": 1.9665130376815796,
+      "learning_rate": 0.0002,
+      "loss": 0.6854,
+      "step": 9410
+    },
+    {
+      "epoch": 7.310826542491269,
+      "grad_norm": 1.9337282180786133,
+      "learning_rate": 0.0002,
+      "loss": 0.6466,
+      "step": 9420
+    },
+    {
+      "epoch": 7.318587504850601,
+      "grad_norm": 1.8582539558410645,
+      "learning_rate": 0.0002,
+      "loss": 0.6885,
+      "step": 9430
+    },
+    {
+      "epoch": 7.326348467209934,
+      "grad_norm": 1.8865947723388672,
+      "learning_rate": 0.0002,
+      "loss": 0.7366,
+      "step": 9440
+    },
+    {
+      "epoch": 7.334109429569267,
+      "grad_norm": 1.8144744634628296,
+      "learning_rate": 0.0002,
+      "loss": 0.7825,
+      "step": 9450
+    },
+    {
+      "epoch": 7.341870391928599,
+      "grad_norm": 1.6930001974105835,
+      "learning_rate": 0.0002,
+      "loss": 0.6706,
+      "step": 9460
+    },
+    {
+      "epoch": 7.349631354287932,
+      "grad_norm": 1.7389107942581177,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 9470
+    },
+    {
+      "epoch": 7.357392316647264,
+      "grad_norm": 1.6860785484313965,
+      "learning_rate": 0.0002,
+      "loss": 0.7264,
+      "step": 9480
+    },
+    {
+      "epoch": 7.365153279006597,
+      "grad_norm": 2.2142202854156494,
+      "learning_rate": 0.0002,
+      "loss": 0.7593,
+      "step": 9490
+    },
+    {
+      "epoch": 7.37291424136593,
+      "grad_norm": 1.9988185167312622,
+      "learning_rate": 0.0002,
+      "loss": 0.7936,
+      "step": 9500
+    },
+    {
+      "epoch": 7.380675203725262,
+      "grad_norm": 1.7517266273498535,
+      "learning_rate": 0.0002,
+      "loss": 0.7348,
+      "step": 9510
+    },
+    {
+      "epoch": 7.3884361660845945,
+      "grad_norm": 1.7426788806915283,
+      "learning_rate": 0.0002,
+      "loss": 0.7432,
+      "step": 9520
+    },
+    {
+      "epoch": 7.396197128443927,
+      "grad_norm": 1.8157157897949219,
+      "learning_rate": 0.0002,
+      "loss": 0.7649,
+      "step": 9530
+    },
+    {
+      "epoch": 7.403958090803259,
+      "grad_norm": 1.9826514720916748,
+      "learning_rate": 0.0002,
+      "loss": 0.716,
+      "step": 9540
+    },
+    {
+      "epoch": 7.411719053162592,
+      "grad_norm": 1.9057488441467285,
+      "learning_rate": 0.0002,
+      "loss": 0.7685,
+      "step": 9550
+    },
+    {
+      "epoch": 7.419480015521925,
+      "grad_norm": 2.0701088905334473,
+      "learning_rate": 0.0002,
+      "loss": 0.7709,
+      "step": 9560
+    },
+    {
+      "epoch": 7.427240977881257,
+      "grad_norm": 1.8623783588409424,
+      "learning_rate": 0.0002,
+      "loss": 0.7438,
+      "step": 9570
+    },
+    {
+      "epoch": 7.43500194024059,
+      "grad_norm": 1.780672550201416,
+      "learning_rate": 0.0002,
+      "loss": 0.7719,
+      "step": 9580
+    },
+    {
+      "epoch": 7.442762902599922,
+      "grad_norm": 1.8437316417694092,
+      "learning_rate": 0.0002,
+      "loss": 0.7579,
+      "step": 9590
+    },
+    {
+      "epoch": 7.450523864959255,
+      "grad_norm": 1.9327329397201538,
+      "learning_rate": 0.0002,
+      "loss": 0.8044,
+      "step": 9600
+    },
+    {
+      "epoch": 7.458284827318588,
+      "grad_norm": 2.011535167694092,
+      "learning_rate": 0.0002,
+      "loss": 0.7212,
+      "step": 9610
+    },
+    {
+      "epoch": 7.46604578967792,
+      "grad_norm": 1.7557756900787354,
+      "learning_rate": 0.0002,
+      "loss": 0.7586,
+      "step": 9620
+    },
+    {
+      "epoch": 7.473806752037253,
+      "grad_norm": 1.8508951663970947,
+      "learning_rate": 0.0002,
+      "loss": 0.7461,
+      "step": 9630
+    },
+    {
+      "epoch": 7.4815677143965855,
+      "grad_norm": 1.8236663341522217,
+      "learning_rate": 0.0002,
+      "loss": 0.7425,
+      "step": 9640
+    },
+    {
+      "epoch": 7.4893286767559175,
+      "grad_norm": 1.9308991432189941,
+      "learning_rate": 0.0002,
+      "loss": 0.7107,
+      "step": 9650
+    },
+    {
+      "epoch": 7.49708963911525,
+      "grad_norm": 1.8095420598983765,
+      "learning_rate": 0.0002,
+      "loss": 0.738,
+      "step": 9660
+    },
+    {
+      "epoch": 7.504850601474583,
+      "grad_norm": 1.8216804265975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7548,
+      "step": 9670
+    },
+    {
+      "epoch": 7.512611563833915,
+      "grad_norm": 1.8275913000106812,
+      "learning_rate": 0.0002,
+      "loss": 0.7494,
+      "step": 9680
+    },
+    {
+      "epoch": 7.520372526193248,
+      "grad_norm": 1.8217864036560059,
+      "learning_rate": 0.0002,
+      "loss": 0.7178,
+      "step": 9690
+    },
+    {
+      "epoch": 7.52813348855258,
+      "grad_norm": 1.461728572845459,
+      "learning_rate": 0.0002,
+      "loss": 0.7331,
+      "step": 9700
+    },
+    {
+      "epoch": 7.535894450911913,
+      "grad_norm": 1.6569337844848633,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 9710
+    },
+    {
+      "epoch": 7.543655413271246,
+      "grad_norm": 2.332101821899414,
+      "learning_rate": 0.0002,
+      "loss": 0.8117,
+      "step": 9720
+    },
+    {
+      "epoch": 7.551416375630578,
+      "grad_norm": 1.760307788848877,
+      "learning_rate": 0.0002,
+      "loss": 0.7937,
+      "step": 9730
+    },
+    {
+      "epoch": 7.559177337989911,
+      "grad_norm": 2.2455198764801025,
+      "learning_rate": 0.0002,
+      "loss": 0.7213,
+      "step": 9740
+    },
+    {
+      "epoch": 7.566938300349244,
+      "grad_norm": 1.819676399230957,
+      "learning_rate": 0.0002,
+      "loss": 0.8122,
+      "step": 9750
+    },
+    {
+      "epoch": 7.574699262708576,
+      "grad_norm": 1.579603910446167,
+      "learning_rate": 0.0002,
+      "loss": 0.754,
+      "step": 9760
+    },
+    {
+      "epoch": 7.5824602250679085,
+      "grad_norm": 1.9687446355819702,
+      "learning_rate": 0.0002,
+      "loss": 0.7608,
+      "step": 9770
+    },
+    {
+      "epoch": 7.590221187427241,
+      "grad_norm": 1.859029769897461,
+      "learning_rate": 0.0002,
+      "loss": 0.813,
+      "step": 9780
+    },
+    {
+      "epoch": 7.597982149786573,
+      "grad_norm": 1.652137279510498,
+      "learning_rate": 0.0002,
+      "loss": 0.7353,
+      "step": 9790
+    },
+    {
+      "epoch": 7.605743112145906,
+      "grad_norm": 1.6381458044052124,
+      "learning_rate": 0.0002,
+      "loss": 0.7388,
+      "step": 9800
+    },
+    {
+      "epoch": 7.613504074505238,
+      "grad_norm": 1.8706856966018677,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 9810
+    },
+    {
+      "epoch": 7.621265036864571,
+      "grad_norm": 1.8709853887557983,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 9820
+    },
+    {
+      "epoch": 7.629025999223904,
+      "grad_norm": 2.1183695793151855,
+      "learning_rate": 0.0002,
+      "loss": 0.7972,
+      "step": 9830
+    },
+    {
+      "epoch": 7.636786961583236,
+      "grad_norm": 1.6770579814910889,
+      "learning_rate": 0.0002,
+      "loss": 0.7645,
+      "step": 9840
+    },
+    {
+      "epoch": 7.644547923942569,
+      "grad_norm": 2.0952677726745605,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 9850
+    },
+    {
+      "epoch": 7.652308886301902,
+      "grad_norm": 1.7378171682357788,
+      "learning_rate": 0.0002,
+      "loss": 0.8483,
+      "step": 9860
+    },
+    {
+      "epoch": 7.660069848661234,
+      "grad_norm": 2.0860157012939453,
+      "learning_rate": 0.0002,
+      "loss": 0.7942,
+      "step": 9870
+    },
+    {
+      "epoch": 7.667830811020567,
+      "grad_norm": 1.6986967325210571,
+      "learning_rate": 0.0002,
+      "loss": 0.769,
+      "step": 9880
+    },
+    {
+      "epoch": 7.6755917733798995,
+      "grad_norm": 1.666702151298523,
+      "learning_rate": 0.0002,
+      "loss": 0.7626,
+      "step": 9890
+    },
+    {
+      "epoch": 7.6833527357392315,
+      "grad_norm": 2.446931838989258,
+      "learning_rate": 0.0002,
+      "loss": 0.7418,
+      "step": 9900
+    },
+    {
+      "epoch": 7.691113698098564,
+      "grad_norm": 1.6984577178955078,
+      "learning_rate": 0.0002,
+      "loss": 0.8333,
+      "step": 9910
+    },
+    {
+      "epoch": 7.698874660457896,
+      "grad_norm": 2.0732316970825195,
+      "learning_rate": 0.0002,
+      "loss": 0.7822,
+      "step": 9920
+    },
+    {
+      "epoch": 7.706635622817229,
+      "grad_norm": 1.9884073734283447,
+      "learning_rate": 0.0002,
+      "loss": 0.7342,
+      "step": 9930
+    },
+    {
+      "epoch": 7.714396585176562,
+      "grad_norm": 1.8669427633285522,
+      "learning_rate": 0.0002,
+      "loss": 0.7825,
+      "step": 9940
+    },
+    {
+      "epoch": 7.722157547535894,
+      "grad_norm": 1.6163996458053589,
+      "learning_rate": 0.0002,
+      "loss": 0.7722,
+      "step": 9950
+    },
+    {
+      "epoch": 7.729918509895227,
+      "grad_norm": 2.4492127895355225,
+      "learning_rate": 0.0002,
+      "loss": 0.8069,
+      "step": 9960
+    },
+    {
+      "epoch": 7.73767947225456,
+      "grad_norm": 1.5625537633895874,
+      "learning_rate": 0.0002,
+      "loss": 0.7924,
+      "step": 9970
+    },
+    {
+      "epoch": 7.745440434613892,
+      "grad_norm": 1.8205251693725586,
+      "learning_rate": 0.0002,
+      "loss": 0.8273,
+      "step": 9980
+    },
+    {
+      "epoch": 7.753201396973225,
+      "grad_norm": 1.9902361631393433,
+      "learning_rate": 0.0002,
+      "loss": 0.7892,
+      "step": 9990
+    },
+    {
+      "epoch": 7.760962359332558,
+      "grad_norm": 1.8294041156768799,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 10000
+    },
+    {
+      "epoch": 7.76872332169189,
+      "grad_norm": 1.8938281536102295,
+      "learning_rate": 0.0002,
+      "loss": 0.771,
+      "step": 10010
+    },
+    {
+      "epoch": 7.7764842840512225,
+      "grad_norm": 1.9367564916610718,
+      "learning_rate": 0.0002,
+      "loss": 0.8304,
+      "step": 10020
+    },
+    {
+      "epoch": 7.7842452464105545,
+      "grad_norm": 1.9130750894546509,
+      "learning_rate": 0.0002,
+      "loss": 0.7437,
+      "step": 10030
+    },
+    {
+      "epoch": 7.792006208769887,
+      "grad_norm": 1.931505799293518,
+      "learning_rate": 0.0002,
+      "loss": 0.8359,
+      "step": 10040
+    },
+    {
+      "epoch": 7.79976717112922,
+      "grad_norm": 2.107954740524292,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 10050
+    },
+    {
+      "epoch": 7.807528133488552,
+      "grad_norm": 1.8736399412155151,
+      "learning_rate": 0.0002,
+      "loss": 0.7922,
+      "step": 10060
+    },
+    {
+      "epoch": 7.815289095847885,
+      "grad_norm": 1.6943566799163818,
+      "learning_rate": 0.0002,
+      "loss": 0.8143,
+      "step": 10070
+    },
+    {
+      "epoch": 7.823050058207218,
+      "grad_norm": 2.365346908569336,
+      "learning_rate": 0.0002,
+      "loss": 0.7645,
+      "step": 10080
+    },
+    {
+      "epoch": 7.83081102056655,
+      "grad_norm": 2.0044665336608887,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 10090
+    },
+    {
+      "epoch": 7.838571982925883,
+      "grad_norm": 1.7680124044418335,
+      "learning_rate": 0.0002,
+      "loss": 0.8297,
+      "step": 10100
+    },
+    {
+      "epoch": 7.846332945285216,
+      "grad_norm": 1.8494547605514526,
+      "learning_rate": 0.0002,
+      "loss": 0.7729,
+      "step": 10110
+    },
+    {
+      "epoch": 7.854093907644548,
+      "grad_norm": 1.822107195854187,
+      "learning_rate": 0.0002,
+      "loss": 0.7909,
+      "step": 10120
+    },
+    {
+      "epoch": 7.861854870003881,
+      "grad_norm": 1.8191527128219604,
+      "learning_rate": 0.0002,
+      "loss": 0.8149,
+      "step": 10130
+    },
+    {
+      "epoch": 7.869615832363213,
+      "grad_norm": 1.8011466264724731,
+      "learning_rate": 0.0002,
+      "loss": 0.8045,
+      "step": 10140
+    },
+    {
+      "epoch": 7.8773767947225455,
+      "grad_norm": 1.7087100744247437,
+      "learning_rate": 0.0002,
+      "loss": 0.79,
+      "step": 10150
+    },
+    {
+      "epoch": 7.885137757081878,
+      "grad_norm": 1.7698721885681152,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 10160
+    },
+    {
+      "epoch": 7.89289871944121,
+      "grad_norm": 1.9578531980514526,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 10170
+    },
+    {
+      "epoch": 7.900659681800543,
+      "grad_norm": 1.7660179138183594,
+      "learning_rate": 0.0002,
+      "loss": 0.8497,
+      "step": 10180
+    },
+    {
+      "epoch": 7.908420644159876,
+      "grad_norm": 2.0122673511505127,
+      "learning_rate": 0.0002,
+      "loss": 0.7858,
+      "step": 10190
+    },
+    {
+      "epoch": 7.916181606519208,
+      "grad_norm": 1.737443208694458,
+      "learning_rate": 0.0002,
+      "loss": 0.7607,
+      "step": 10200
+    },
+    {
+      "epoch": 7.923942568878541,
+      "grad_norm": 1.6381052732467651,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 10210
+    },
+    {
+      "epoch": 7.931703531237874,
+      "grad_norm": 1.8845038414001465,
+      "learning_rate": 0.0002,
+      "loss": 0.7855,
+      "step": 10220
+    },
+    {
+      "epoch": 7.939464493597206,
+      "grad_norm": 1.952194333076477,
+      "learning_rate": 0.0002,
+      "loss": 0.8354,
+      "step": 10230
+    },
+    {
+      "epoch": 7.947225455956539,
+      "grad_norm": 1.7254410982131958,
+      "learning_rate": 0.0002,
+      "loss": 0.8428,
+      "step": 10240
+    },
+    {
+      "epoch": 7.954986418315871,
+      "grad_norm": 2.14776873588562,
+      "learning_rate": 0.0002,
+      "loss": 0.834,
+      "step": 10250
+    },
+    {
+      "epoch": 7.962747380675204,
+      "grad_norm": 1.7655725479125977,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 10260
+    },
+    {
+      "epoch": 7.9705083430345365,
+      "grad_norm": 1.7337331771850586,
+      "learning_rate": 0.0002,
+      "loss": 0.8176,
+      "step": 10270
+    },
+    {
+      "epoch": 7.9782693053938685,
+      "grad_norm": 1.7742228507995605,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 10280
+    },
+    {
+      "epoch": 7.986030267753201,
+      "grad_norm": 1.892137050628662,
+      "learning_rate": 0.0002,
+      "loss": 0.7788,
+      "step": 10290
+    },
+    {
+      "epoch": 7.993791230112534,
+      "grad_norm": 1.8636300563812256,
+      "learning_rate": 0.0002,
+      "loss": 0.8243,
+      "step": 10300
+    },
+    {
+      "epoch": 7.996895615056267,
+      "eval_loss": 2.7464993000030518,
+      "eval_runtime": 114.078,
+      "eval_samples_per_second": 4.444,
+      "eval_steps_per_second": 0.561,
+      "step": 10304
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.2925203495256064e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-10304/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5208229e29fb38c73343f46a72a64976ddf01127
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a18343e5c8a561cfe90c00254345da8a62839392f1f92154472ccfeac3f0492e
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f7e289d6e4092c63f1963d86dd2f62e0ff6e5487
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67873c50dcb6182b8741740688f5c1098ab9b75ddd7a223f632ceffef2b831c2
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c0365d362b90a2c12afa3608502a350b8bec9b5a
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85fdc4271a5e20d6f1f0dd445a31121166a83dd6ab00011bf0901a2ca8c5c7a2
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8a59812c3593eca842c322aba5a497cd945702bc
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a17393a475a6079ffa2869cfabd4d1a282c203c4adf92772c840de5a59a54b6b
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..e3d851b24912a599aea296ec87d2e80723886f86
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/trainer_state.json
@@ -0,0 +1,937 @@
+{
+  "best_metric": 1.8081045150756836,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288",
+  "epoch": 0.9996119518820333,
+  "eval_steps": 10,
+  "global_step": 1288,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 6.618218624188416e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6736da95acf74ac199158de2bcd2f45e279c6a48
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38b9d4be16ebe4fd6dec61f83710d425e98161ba60be90a89bdaeb9c11beb3ef
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..955f398d8a589f6bb4ef6b0e46582f5db08a524a
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:065e8ea19f402b886f338dbbcfa06cd1dcb161bbbae9716a0d68d69f6a74cfe2
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..210ece09eb39d86bcfd971c1e9bfbaec3f4f12d0
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:366694b55f261d0313ae3ec388be059c6630cdad34deff87df96e95222d25afb
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b73d0bd92d62d986603cc11993cc9270ad145f90
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4358b9377fa65c761c44392416e6d2355af889e9461d685fe4753ef6a2cf58e
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..603844c8e3f1683050ba1437515d04cc5238909d
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/trainer_state.json
@@ -0,0 +1,1848 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2577,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 1.3236437248376832e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2786b01ff7eed83657f3076910e481f97a6b9d97
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f51b151c750ccb60ec370ef0a0102921296969c48183896477de09e0fad7ae4
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..148f058292122a4256bf08a42124876ee21246fa
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82e48ecfc9ee6499497242616ba9c97813a50711664bfef1f982ff4a5b773e01
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f8ca8dba68bda9a461f59de0c7c96e8f29fb78ad
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc96eed9e4fc239cc5ce62a654c9651acec2fda09e9f5ff69a92e5f6c0c9741c
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e88deae00cb5a1806f83d7e94c6d37bada812744
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b2ce4676fa126c497a7bdef3c433de52a10b5a4190bb774d68ef751fa2b4e2c
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5baed5ecb767c4b2e4ce4448030514f3427f5919
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/trainer_state.json
@@ -0,0 +1,2759 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 2.9996119518820334,
+  "eval_steps": 10,
+  "global_step": 3865,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 1.9854655872565248e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-3865/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fcf639104461207a045d5d09886005635fa03687
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcc71afcf27414c77213ed367ce97289db2b15abf576c01b46ddfd036010587e
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ddcf3fe5f911ac00a4ba3b14a6dc5f361a40aeb9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6bbf02e1091f7198aaf856f46cbc13189efaa221eb7f4ebd80267834cafaa91
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6c947ffdd2513b180e20aaa51ecd959f7d3624ae
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a9146ec642d2e9f5ee7472f453c10c017aa04a0cdfeede141c07559c9843082
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ba216a635b5168dc97e9028e9895f5527ec74cf
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56694862b165f53e8e924e92cff89c060ce3f55cea71f7162761fb9876c8b672
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7c8530012effdcba26024d9583116b824efecb5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/trainer_state.json
@@ -0,0 +1,3670 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5154,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    },
+    {
+      "epoch": 3.0034924330616994,
+      "grad_norm": 0.5645653605461121,
+      "learning_rate": 0.0002,
+      "loss": 1.4983,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0112533954210323,
+      "grad_norm": 0.6457151174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0190143577803648,
+      "grad_norm": 0.583838164806366,
+      "learning_rate": 0.0002,
+      "loss": 1.3899,
+      "step": 3890
+    },
+    {
+      "epoch": 3.026775320139697,
+      "grad_norm": 0.6819260120391846,
+      "learning_rate": 0.0002,
+      "loss": 1.3258,
+      "step": 3900
+    },
+    {
+      "epoch": 3.03453628249903,
+      "grad_norm": 0.6692903637886047,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0422972448583625,
+      "grad_norm": 0.6101024746894836,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 3920
+    },
+    {
+      "epoch": 3.050058207217695,
+      "grad_norm": 0.7014093399047852,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3930
+    },
+    {
+      "epoch": 3.0578191695770274,
+      "grad_norm": 0.7380381226539612,
+      "learning_rate": 0.0002,
+      "loss": 1.3885,
+      "step": 3940
+    },
+    {
+      "epoch": 3.0655801319363603,
+      "grad_norm": 0.6607900857925415,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0733410942956927,
+      "grad_norm": 0.735263466835022,
+      "learning_rate": 0.0002,
+      "loss": 1.4293,
+      "step": 3960
+    },
+    {
+      "epoch": 3.081102056655025,
+      "grad_norm": 0.6788513660430908,
+      "learning_rate": 0.0002,
+      "loss": 1.3966,
+      "step": 3970
+    },
+    {
+      "epoch": 3.088863019014358,
+      "grad_norm": 0.6347652673721313,
+      "learning_rate": 0.0002,
+      "loss": 1.3435,
+      "step": 3980
+    },
+    {
+      "epoch": 3.0966239813736904,
+      "grad_norm": 0.7056642770767212,
+      "learning_rate": 0.0002,
+      "loss": 1.4518,
+      "step": 3990
+    },
+    {
+      "epoch": 3.104384943733023,
+      "grad_norm": 0.6387075185775757,
+      "learning_rate": 0.0002,
+      "loss": 1.4474,
+      "step": 4000
+    },
+    {
+      "epoch": 3.1121459060923553,
+      "grad_norm": 0.6701116561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.3833,
+      "step": 4010
+    },
+    {
+      "epoch": 3.119906868451688,
+      "grad_norm": 0.7558449506759644,
+      "learning_rate": 0.0002,
+      "loss": 1.404,
+      "step": 4020
+    },
+    {
+      "epoch": 3.1276678308110206,
+      "grad_norm": 0.6612881422042847,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 4030
+    },
+    {
+      "epoch": 3.135428793170353,
+      "grad_norm": 0.7474587559700012,
+      "learning_rate": 0.0002,
+      "loss": 1.439,
+      "step": 4040
+    },
+    {
+      "epoch": 3.1431897555296855,
+      "grad_norm": 0.7292373776435852,
+      "learning_rate": 0.0002,
+      "loss": 1.4616,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1509507178890184,
+      "grad_norm": 0.7432886958122253,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 4060
+    },
+    {
+      "epoch": 3.158711680248351,
+      "grad_norm": 0.6366098523139954,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1664726426076832,
+      "grad_norm": 0.6837611794471741,
+      "learning_rate": 0.0002,
+      "loss": 1.5044,
+      "step": 4080
+    },
+    {
+      "epoch": 3.174233604967016,
+      "grad_norm": 0.7194393277168274,
+      "learning_rate": 0.0002,
+      "loss": 1.4332,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1819945673263486,
+      "grad_norm": 0.6963607668876648,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 4100
+    },
+    {
+      "epoch": 3.189755529685681,
+      "grad_norm": 0.6404902935028076,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4110
+    },
+    {
+      "epoch": 3.1975164920450134,
+      "grad_norm": 0.7172070741653442,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 4120
+    },
+    {
+      "epoch": 3.2052774544043463,
+      "grad_norm": 0.6577759385108948,
+      "learning_rate": 0.0002,
+      "loss": 1.4658,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2130384167636787,
+      "grad_norm": 0.6658480167388916,
+      "learning_rate": 0.0002,
+      "loss": 1.4019,
+      "step": 4140
+    },
+    {
+      "epoch": 3.220799379123011,
+      "grad_norm": 0.6771699786186218,
+      "learning_rate": 0.0002,
+      "loss": 1.4348,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2285603414823436,
+      "grad_norm": 0.699035108089447,
+      "learning_rate": 0.0002,
+      "loss": 1.4736,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2363213038416765,
+      "grad_norm": 0.7218514680862427,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 4170
+    },
+    {
+      "epoch": 3.244082266201009,
+      "grad_norm": 0.6270631551742554,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 4180
+    },
+    {
+      "epoch": 3.2518432285603414,
+      "grad_norm": 0.6828921437263489,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2596041909196742,
+      "grad_norm": 0.6005498170852661,
+      "learning_rate": 0.0002,
+      "loss": 1.4663,
+      "step": 4200
+    },
+    {
+      "epoch": 3.2673651532790067,
+      "grad_norm": 0.6974790692329407,
+      "learning_rate": 0.0002,
+      "loss": 1.4798,
+      "step": 4210
+    },
+    {
+      "epoch": 3.275126115638339,
+      "grad_norm": 0.7269543409347534,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4220
+    },
+    {
+      "epoch": 3.2828870779976715,
+      "grad_norm": 0.6728787422180176,
+      "learning_rate": 0.0002,
+      "loss": 1.3848,
+      "step": 4230
+    },
+    {
+      "epoch": 3.2906480403570044,
+      "grad_norm": 0.676972508430481,
+      "learning_rate": 0.0002,
+      "loss": 1.4112,
+      "step": 4240
+    },
+    {
+      "epoch": 3.298409002716337,
+      "grad_norm": 0.748309314250946,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3061699650756693,
+      "grad_norm": 0.6976589560508728,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 4260
+    },
+    {
+      "epoch": 3.3139309274350017,
+      "grad_norm": 0.649780809879303,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3216918897943346,
+      "grad_norm": 0.6529902815818787,
+      "learning_rate": 0.0002,
+      "loss": 1.327,
+      "step": 4280
+    },
+    {
+      "epoch": 3.329452852153667,
+      "grad_norm": 0.9273163676261902,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 4290
+    },
+    {
+      "epoch": 3.3372138145129995,
+      "grad_norm": 0.717024028301239,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 4300
+    },
+    {
+      "epoch": 3.3449747768723324,
+      "grad_norm": 0.7914950251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.4441,
+      "step": 4310
+    },
+    {
+      "epoch": 3.352735739231665,
+      "grad_norm": 0.7133203148841858,
+      "learning_rate": 0.0002,
+      "loss": 1.432,
+      "step": 4320
+    },
+    {
+      "epoch": 3.3604967015909972,
+      "grad_norm": 0.7409568428993225,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3682576639503297,
+      "grad_norm": 0.6993981003761292,
+      "learning_rate": 0.0002,
+      "loss": 1.3992,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3760186263096625,
+      "grad_norm": 0.7114535570144653,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4350
+    },
+    {
+      "epoch": 3.383779588668995,
+      "grad_norm": 0.6790860295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.4227,
+      "step": 4360
+    },
+    {
+      "epoch": 3.3915405510283274,
+      "grad_norm": 0.6507849097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 4370
+    },
+    {
+      "epoch": 3.39930151338766,
+      "grad_norm": 0.5967804193496704,
+      "learning_rate": 0.0002,
+      "loss": 1.4559,
+      "step": 4380
+    },
+    {
+      "epoch": 3.4070624757469927,
+      "grad_norm": 0.6625847816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 4390
+    },
+    {
+      "epoch": 3.414823438106325,
+      "grad_norm": 0.6736508011817932,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4225844004656576,
+      "grad_norm": 0.7870860695838928,
+      "learning_rate": 0.0002,
+      "loss": 1.4363,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4303453628249905,
+      "grad_norm": 0.7205295562744141,
+      "learning_rate": 0.0002,
+      "loss": 1.4114,
+      "step": 4420
+    },
+    {
+      "epoch": 3.438106325184323,
+      "grad_norm": 0.6634634137153625,
+      "learning_rate": 0.0002,
+      "loss": 1.4131,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4458672875436553,
+      "grad_norm": 0.7562733292579651,
+      "learning_rate": 0.0002,
+      "loss": 1.4683,
+      "step": 4440
+    },
+    {
+      "epoch": 3.453628249902988,
+      "grad_norm": 0.6585879921913147,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4613892122623207,
+      "grad_norm": 0.6896792054176331,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 4460
+    },
+    {
+      "epoch": 3.469150174621653,
+      "grad_norm": 0.6520342230796814,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4769111369809855,
+      "grad_norm": 0.6760806441307068,
+      "learning_rate": 0.0002,
+      "loss": 1.3423,
+      "step": 4480
+    },
+    {
+      "epoch": 3.484672099340318,
+      "grad_norm": 0.7539774179458618,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 4490
+    },
+    {
+      "epoch": 3.492433061699651,
+      "grad_norm": 0.7409411668777466,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 4500
+    },
+    {
+      "epoch": 3.5001940240589833,
+      "grad_norm": 0.6876253485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.4069,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5079549864183157,
+      "grad_norm": 0.7028461694717407,
+      "learning_rate": 0.0002,
+      "loss": 1.4228,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5157159487776486,
+      "grad_norm": 0.8056529760360718,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4530
+    },
+    {
+      "epoch": 3.523476911136981,
+      "grad_norm": 0.711338996887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5312378734963135,
+      "grad_norm": 0.7343552708625793,
+      "learning_rate": 0.0002,
+      "loss": 1.5247,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5389988358556463,
+      "grad_norm": 0.745479941368103,
+      "learning_rate": 0.0002,
+      "loss": 1.4308,
+      "step": 4560
+    },
+    {
+      "epoch": 3.5467597982149788,
+      "grad_norm": 0.7582294940948486,
+      "learning_rate": 0.0002,
+      "loss": 1.4229,
+      "step": 4570
+    },
+    {
+      "epoch": 3.554520760574311,
+      "grad_norm": 0.6717444658279419,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4580
+    },
+    {
+      "epoch": 3.5622817229336436,
+      "grad_norm": 0.7417883276939392,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 4590
+    },
+    {
+      "epoch": 3.570042685292976,
+      "grad_norm": 0.6385737061500549,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4600
+    },
+    {
+      "epoch": 3.577803647652309,
+      "grad_norm": 0.716704249382019,
+      "learning_rate": 0.0002,
+      "loss": 1.3981,
+      "step": 4610
+    },
+    {
+      "epoch": 3.5855646100116414,
+      "grad_norm": 0.6948980093002319,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 4620
+    },
+    {
+      "epoch": 3.593325572370974,
+      "grad_norm": 0.6961140036582947,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6010865347303067,
+      "grad_norm": 0.7493122220039368,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 4640
+    },
+    {
+      "epoch": 3.608847497089639,
+      "grad_norm": 0.7431658506393433,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6166084594489716,
+      "grad_norm": 0.8353387713432312,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4660
+    },
+    {
+      "epoch": 3.6243694218083045,
+      "grad_norm": 0.7095612287521362,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 4670
+    },
+    {
+      "epoch": 3.632130384167637,
+      "grad_norm": 0.776620090007782,
+      "learning_rate": 0.0002,
+      "loss": 1.4003,
+      "step": 4680
+    },
+    {
+      "epoch": 3.6398913465269693,
+      "grad_norm": 0.7198925018310547,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6476523088863018,
+      "grad_norm": 0.8238834738731384,
+      "learning_rate": 0.0002,
+      "loss": 1.4543,
+      "step": 4700
+    },
+    {
+      "epoch": 3.655413271245634,
+      "grad_norm": 0.6804245710372925,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4710
+    },
+    {
+      "epoch": 3.663174233604967,
+      "grad_norm": 0.8444845676422119,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 4720
+    },
+    {
+      "epoch": 3.6709351959642995,
+      "grad_norm": 0.743797779083252,
+      "learning_rate": 0.0002,
+      "loss": 1.3825,
+      "step": 4730
+    },
+    {
+      "epoch": 3.678696158323632,
+      "grad_norm": 0.8994188904762268,
+      "learning_rate": 0.0002,
+      "loss": 1.4213,
+      "step": 4740
+    },
+    {
+      "epoch": 3.686457120682965,
+      "grad_norm": 0.75416100025177,
+      "learning_rate": 0.0002,
+      "loss": 1.4281,
+      "step": 4750
+    },
+    {
+      "epoch": 3.6942180830422973,
+      "grad_norm": 0.6499266028404236,
+      "learning_rate": 0.0002,
+      "loss": 1.4154,
+      "step": 4760
+    },
+    {
+      "epoch": 3.7019790454016297,
+      "grad_norm": 0.7246791124343872,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7097400077609626,
+      "grad_norm": 0.7831124067306519,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 4780
+    },
+    {
+      "epoch": 3.717500970120295,
+      "grad_norm": 0.7130028009414673,
+      "learning_rate": 0.0002,
+      "loss": 1.3933,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7252619324796274,
+      "grad_norm": 0.7501602172851562,
+      "learning_rate": 0.0002,
+      "loss": 1.4632,
+      "step": 4800
+    },
+    {
+      "epoch": 3.73302289483896,
+      "grad_norm": 0.6980932950973511,
+      "learning_rate": 0.0002,
+      "loss": 1.4985,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7407838571982923,
+      "grad_norm": 0.8050530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.4517,
+      "step": 4820
+    },
+    {
+      "epoch": 3.748544819557625,
+      "grad_norm": 0.6385579705238342,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7563057819169576,
+      "grad_norm": 0.6664714813232422,
+      "learning_rate": 0.0002,
+      "loss": 1.5281,
+      "step": 4840
+    },
+    {
+      "epoch": 3.76406674427629,
+      "grad_norm": 0.7125676274299622,
+      "learning_rate": 0.0002,
+      "loss": 1.4443,
+      "step": 4850
+    },
+    {
+      "epoch": 3.771827706635623,
+      "grad_norm": 0.7231866717338562,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4860
+    },
+    {
+      "epoch": 3.7795886689949554,
+      "grad_norm": 0.6917183995246887,
+      "learning_rate": 0.0002,
+      "loss": 1.4446,
+      "step": 4870
+    },
+    {
+      "epoch": 3.787349631354288,
+      "grad_norm": 0.665037989616394,
+      "learning_rate": 0.0002,
+      "loss": 1.4369,
+      "step": 4880
+    },
+    {
+      "epoch": 3.7951105937136207,
+      "grad_norm": 0.5837726593017578,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4890
+    },
+    {
+      "epoch": 3.802871556072953,
+      "grad_norm": 0.6366701722145081,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8106325184322856,
+      "grad_norm": 0.7082223892211914,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 4910
+    },
+    {
+      "epoch": 3.818393480791618,
+      "grad_norm": 0.8101672530174255,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4920
+    },
+    {
+      "epoch": 3.826154443150951,
+      "grad_norm": 0.7516148090362549,
+      "learning_rate": 0.0002,
+      "loss": 1.3659,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8339154055102833,
+      "grad_norm": 0.7928489446640015,
+      "learning_rate": 0.0002,
+      "loss": 1.3909,
+      "step": 4940
+    },
+    {
+      "epoch": 3.8416763678696157,
+      "grad_norm": 0.6892234683036804,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 4950
+    },
+    {
+      "epoch": 3.849437330228948,
+      "grad_norm": 0.6381304264068604,
+      "learning_rate": 0.0002,
+      "loss": 1.5024,
+      "step": 4960
+    },
+    {
+      "epoch": 3.857198292588281,
+      "grad_norm": 0.8068831562995911,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8649592549476135,
+      "grad_norm": 0.7289869785308838,
+      "learning_rate": 0.0002,
+      "loss": 1.45,
+      "step": 4980
+    },
+    {
+      "epoch": 3.872720217306946,
+      "grad_norm": 0.7278549075126648,
+      "learning_rate": 0.0002,
+      "loss": 1.398,
+      "step": 4990
+    },
+    {
+      "epoch": 3.880481179666279,
+      "grad_norm": 0.7324236631393433,
+      "learning_rate": 0.0002,
+      "loss": 1.4442,
+      "step": 5000
+    },
+    {
+      "epoch": 3.8882421420256112,
+      "grad_norm": 0.6759871244430542,
+      "learning_rate": 0.0002,
+      "loss": 1.4511,
+      "step": 5010
+    },
+    {
+      "epoch": 3.8960031043849437,
+      "grad_norm": 0.8159207701683044,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 5020
+    },
+    {
+      "epoch": 3.9037640667442766,
+      "grad_norm": 0.6536211967468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5030
+    },
+    {
+      "epoch": 3.911525029103609,
+      "grad_norm": 0.6827932000160217,
+      "learning_rate": 0.0002,
+      "loss": 1.4335,
+      "step": 5040
+    },
+    {
+      "epoch": 3.9192859914629414,
+      "grad_norm": 0.6688340306282043,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 5050
+    },
+    {
+      "epoch": 3.927046953822274,
+      "grad_norm": 0.6385695934295654,
+      "learning_rate": 0.0002,
+      "loss": 1.4099,
+      "step": 5060
+    },
+    {
+      "epoch": 3.9348079161816063,
+      "grad_norm": 0.6975107192993164,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 5070
+    },
+    {
+      "epoch": 3.942568878540939,
+      "grad_norm": 0.6684112548828125,
+      "learning_rate": 0.0002,
+      "loss": 1.4893,
+      "step": 5080
+    },
+    {
+      "epoch": 3.9503298409002716,
+      "grad_norm": 0.8349628448486328,
+      "learning_rate": 0.0002,
+      "loss": 1.4732,
+      "step": 5090
+    },
+    {
+      "epoch": 3.958090803259604,
+      "grad_norm": 0.7146425843238831,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 5100
+    },
+    {
+      "epoch": 3.965851765618937,
+      "grad_norm": 0.6555036902427673,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9736127279782694,
+      "grad_norm": 0.7037415504455566,
+      "learning_rate": 0.0002,
+      "loss": 1.4274,
+      "step": 5120
+    },
+    {
+      "epoch": 3.981373690337602,
+      "grad_norm": 0.7235575914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5130
+    },
+    {
+      "epoch": 3.9891346526969347,
+      "grad_norm": 0.7092325687408447,
+      "learning_rate": 0.0002,
+      "loss": 1.4455,
+      "step": 5140
+    },
+    {
+      "epoch": 3.996895615056267,
+      "grad_norm": 0.7490319609642029,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 5150
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9131355285644531,
+      "eval_runtime": 105.5778,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 5154
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 2.6472874496753664e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-5154/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..abe4ebc4beb969d5267a45992ec656a4a68b1fd9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e402faa5f84824ca939ad7b1591b51dd6860ad6548e9197d4604e851281827f5
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..626a7ec7a3016789a3fc36fdf676584a6548c853
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16e2d815ba587c0cc13e35c2f1ab31b4cf7397e0816df4838bbadfb4acc88060
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..af8ad5825e9f12e91d91c25a57b7f1946b85b72e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a1f14aee1403446037fc0dbcc8de020d13071880a931dead057e5c41bed610f
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4e4b23fd9d976993d82932d2cd457286e4531bcc
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff08e7f23e0e79786a6547122bbe0f72d7de03e3cf2d4a4088c28046431d6b1
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..ef5191874f9128d31f7e901f5f6178901b2d1c15
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/trainer_state.json
@@ -0,0 +1,4581 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 4.9996119518820334,
+  "eval_steps": 10,
+  "global_step": 6442,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    },
+    {
+      "epoch": 3.0034924330616994,
+      "grad_norm": 0.5645653605461121,
+      "learning_rate": 0.0002,
+      "loss": 1.4983,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0112533954210323,
+      "grad_norm": 0.6457151174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0190143577803648,
+      "grad_norm": 0.583838164806366,
+      "learning_rate": 0.0002,
+      "loss": 1.3899,
+      "step": 3890
+    },
+    {
+      "epoch": 3.026775320139697,
+      "grad_norm": 0.6819260120391846,
+      "learning_rate": 0.0002,
+      "loss": 1.3258,
+      "step": 3900
+    },
+    {
+      "epoch": 3.03453628249903,
+      "grad_norm": 0.6692903637886047,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0422972448583625,
+      "grad_norm": 0.6101024746894836,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 3920
+    },
+    {
+      "epoch": 3.050058207217695,
+      "grad_norm": 0.7014093399047852,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3930
+    },
+    {
+      "epoch": 3.0578191695770274,
+      "grad_norm": 0.7380381226539612,
+      "learning_rate": 0.0002,
+      "loss": 1.3885,
+      "step": 3940
+    },
+    {
+      "epoch": 3.0655801319363603,
+      "grad_norm": 0.6607900857925415,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0733410942956927,
+      "grad_norm": 0.735263466835022,
+      "learning_rate": 0.0002,
+      "loss": 1.4293,
+      "step": 3960
+    },
+    {
+      "epoch": 3.081102056655025,
+      "grad_norm": 0.6788513660430908,
+      "learning_rate": 0.0002,
+      "loss": 1.3966,
+      "step": 3970
+    },
+    {
+      "epoch": 3.088863019014358,
+      "grad_norm": 0.6347652673721313,
+      "learning_rate": 0.0002,
+      "loss": 1.3435,
+      "step": 3980
+    },
+    {
+      "epoch": 3.0966239813736904,
+      "grad_norm": 0.7056642770767212,
+      "learning_rate": 0.0002,
+      "loss": 1.4518,
+      "step": 3990
+    },
+    {
+      "epoch": 3.104384943733023,
+      "grad_norm": 0.6387075185775757,
+      "learning_rate": 0.0002,
+      "loss": 1.4474,
+      "step": 4000
+    },
+    {
+      "epoch": 3.1121459060923553,
+      "grad_norm": 0.6701116561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.3833,
+      "step": 4010
+    },
+    {
+      "epoch": 3.119906868451688,
+      "grad_norm": 0.7558449506759644,
+      "learning_rate": 0.0002,
+      "loss": 1.404,
+      "step": 4020
+    },
+    {
+      "epoch": 3.1276678308110206,
+      "grad_norm": 0.6612881422042847,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 4030
+    },
+    {
+      "epoch": 3.135428793170353,
+      "grad_norm": 0.7474587559700012,
+      "learning_rate": 0.0002,
+      "loss": 1.439,
+      "step": 4040
+    },
+    {
+      "epoch": 3.1431897555296855,
+      "grad_norm": 0.7292373776435852,
+      "learning_rate": 0.0002,
+      "loss": 1.4616,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1509507178890184,
+      "grad_norm": 0.7432886958122253,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 4060
+    },
+    {
+      "epoch": 3.158711680248351,
+      "grad_norm": 0.6366098523139954,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1664726426076832,
+      "grad_norm": 0.6837611794471741,
+      "learning_rate": 0.0002,
+      "loss": 1.5044,
+      "step": 4080
+    },
+    {
+      "epoch": 3.174233604967016,
+      "grad_norm": 0.7194393277168274,
+      "learning_rate": 0.0002,
+      "loss": 1.4332,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1819945673263486,
+      "grad_norm": 0.6963607668876648,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 4100
+    },
+    {
+      "epoch": 3.189755529685681,
+      "grad_norm": 0.6404902935028076,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4110
+    },
+    {
+      "epoch": 3.1975164920450134,
+      "grad_norm": 0.7172070741653442,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 4120
+    },
+    {
+      "epoch": 3.2052774544043463,
+      "grad_norm": 0.6577759385108948,
+      "learning_rate": 0.0002,
+      "loss": 1.4658,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2130384167636787,
+      "grad_norm": 0.6658480167388916,
+      "learning_rate": 0.0002,
+      "loss": 1.4019,
+      "step": 4140
+    },
+    {
+      "epoch": 3.220799379123011,
+      "grad_norm": 0.6771699786186218,
+      "learning_rate": 0.0002,
+      "loss": 1.4348,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2285603414823436,
+      "grad_norm": 0.699035108089447,
+      "learning_rate": 0.0002,
+      "loss": 1.4736,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2363213038416765,
+      "grad_norm": 0.7218514680862427,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 4170
+    },
+    {
+      "epoch": 3.244082266201009,
+      "grad_norm": 0.6270631551742554,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 4180
+    },
+    {
+      "epoch": 3.2518432285603414,
+      "grad_norm": 0.6828921437263489,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2596041909196742,
+      "grad_norm": 0.6005498170852661,
+      "learning_rate": 0.0002,
+      "loss": 1.4663,
+      "step": 4200
+    },
+    {
+      "epoch": 3.2673651532790067,
+      "grad_norm": 0.6974790692329407,
+      "learning_rate": 0.0002,
+      "loss": 1.4798,
+      "step": 4210
+    },
+    {
+      "epoch": 3.275126115638339,
+      "grad_norm": 0.7269543409347534,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4220
+    },
+    {
+      "epoch": 3.2828870779976715,
+      "grad_norm": 0.6728787422180176,
+      "learning_rate": 0.0002,
+      "loss": 1.3848,
+      "step": 4230
+    },
+    {
+      "epoch": 3.2906480403570044,
+      "grad_norm": 0.676972508430481,
+      "learning_rate": 0.0002,
+      "loss": 1.4112,
+      "step": 4240
+    },
+    {
+      "epoch": 3.298409002716337,
+      "grad_norm": 0.748309314250946,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3061699650756693,
+      "grad_norm": 0.6976589560508728,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 4260
+    },
+    {
+      "epoch": 3.3139309274350017,
+      "grad_norm": 0.649780809879303,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3216918897943346,
+      "grad_norm": 0.6529902815818787,
+      "learning_rate": 0.0002,
+      "loss": 1.327,
+      "step": 4280
+    },
+    {
+      "epoch": 3.329452852153667,
+      "grad_norm": 0.9273163676261902,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 4290
+    },
+    {
+      "epoch": 3.3372138145129995,
+      "grad_norm": 0.717024028301239,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 4300
+    },
+    {
+      "epoch": 3.3449747768723324,
+      "grad_norm": 0.7914950251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.4441,
+      "step": 4310
+    },
+    {
+      "epoch": 3.352735739231665,
+      "grad_norm": 0.7133203148841858,
+      "learning_rate": 0.0002,
+      "loss": 1.432,
+      "step": 4320
+    },
+    {
+      "epoch": 3.3604967015909972,
+      "grad_norm": 0.7409568428993225,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3682576639503297,
+      "grad_norm": 0.6993981003761292,
+      "learning_rate": 0.0002,
+      "loss": 1.3992,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3760186263096625,
+      "grad_norm": 0.7114535570144653,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4350
+    },
+    {
+      "epoch": 3.383779588668995,
+      "grad_norm": 0.6790860295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.4227,
+      "step": 4360
+    },
+    {
+      "epoch": 3.3915405510283274,
+      "grad_norm": 0.6507849097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 4370
+    },
+    {
+      "epoch": 3.39930151338766,
+      "grad_norm": 0.5967804193496704,
+      "learning_rate": 0.0002,
+      "loss": 1.4559,
+      "step": 4380
+    },
+    {
+      "epoch": 3.4070624757469927,
+      "grad_norm": 0.6625847816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 4390
+    },
+    {
+      "epoch": 3.414823438106325,
+      "grad_norm": 0.6736508011817932,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4225844004656576,
+      "grad_norm": 0.7870860695838928,
+      "learning_rate": 0.0002,
+      "loss": 1.4363,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4303453628249905,
+      "grad_norm": 0.7205295562744141,
+      "learning_rate": 0.0002,
+      "loss": 1.4114,
+      "step": 4420
+    },
+    {
+      "epoch": 3.438106325184323,
+      "grad_norm": 0.6634634137153625,
+      "learning_rate": 0.0002,
+      "loss": 1.4131,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4458672875436553,
+      "grad_norm": 0.7562733292579651,
+      "learning_rate": 0.0002,
+      "loss": 1.4683,
+      "step": 4440
+    },
+    {
+      "epoch": 3.453628249902988,
+      "grad_norm": 0.6585879921913147,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4613892122623207,
+      "grad_norm": 0.6896792054176331,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 4460
+    },
+    {
+      "epoch": 3.469150174621653,
+      "grad_norm": 0.6520342230796814,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4769111369809855,
+      "grad_norm": 0.6760806441307068,
+      "learning_rate": 0.0002,
+      "loss": 1.3423,
+      "step": 4480
+    },
+    {
+      "epoch": 3.484672099340318,
+      "grad_norm": 0.7539774179458618,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 4490
+    },
+    {
+      "epoch": 3.492433061699651,
+      "grad_norm": 0.7409411668777466,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 4500
+    },
+    {
+      "epoch": 3.5001940240589833,
+      "grad_norm": 0.6876253485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.4069,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5079549864183157,
+      "grad_norm": 0.7028461694717407,
+      "learning_rate": 0.0002,
+      "loss": 1.4228,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5157159487776486,
+      "grad_norm": 0.8056529760360718,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4530
+    },
+    {
+      "epoch": 3.523476911136981,
+      "grad_norm": 0.711338996887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5312378734963135,
+      "grad_norm": 0.7343552708625793,
+      "learning_rate": 0.0002,
+      "loss": 1.5247,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5389988358556463,
+      "grad_norm": 0.745479941368103,
+      "learning_rate": 0.0002,
+      "loss": 1.4308,
+      "step": 4560
+    },
+    {
+      "epoch": 3.5467597982149788,
+      "grad_norm": 0.7582294940948486,
+      "learning_rate": 0.0002,
+      "loss": 1.4229,
+      "step": 4570
+    },
+    {
+      "epoch": 3.554520760574311,
+      "grad_norm": 0.6717444658279419,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4580
+    },
+    {
+      "epoch": 3.5622817229336436,
+      "grad_norm": 0.7417883276939392,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 4590
+    },
+    {
+      "epoch": 3.570042685292976,
+      "grad_norm": 0.6385737061500549,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4600
+    },
+    {
+      "epoch": 3.577803647652309,
+      "grad_norm": 0.716704249382019,
+      "learning_rate": 0.0002,
+      "loss": 1.3981,
+      "step": 4610
+    },
+    {
+      "epoch": 3.5855646100116414,
+      "grad_norm": 0.6948980093002319,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 4620
+    },
+    {
+      "epoch": 3.593325572370974,
+      "grad_norm": 0.6961140036582947,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6010865347303067,
+      "grad_norm": 0.7493122220039368,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 4640
+    },
+    {
+      "epoch": 3.608847497089639,
+      "grad_norm": 0.7431658506393433,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6166084594489716,
+      "grad_norm": 0.8353387713432312,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4660
+    },
+    {
+      "epoch": 3.6243694218083045,
+      "grad_norm": 0.7095612287521362,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 4670
+    },
+    {
+      "epoch": 3.632130384167637,
+      "grad_norm": 0.776620090007782,
+      "learning_rate": 0.0002,
+      "loss": 1.4003,
+      "step": 4680
+    },
+    {
+      "epoch": 3.6398913465269693,
+      "grad_norm": 0.7198925018310547,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6476523088863018,
+      "grad_norm": 0.8238834738731384,
+      "learning_rate": 0.0002,
+      "loss": 1.4543,
+      "step": 4700
+    },
+    {
+      "epoch": 3.655413271245634,
+      "grad_norm": 0.6804245710372925,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4710
+    },
+    {
+      "epoch": 3.663174233604967,
+      "grad_norm": 0.8444845676422119,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 4720
+    },
+    {
+      "epoch": 3.6709351959642995,
+      "grad_norm": 0.743797779083252,
+      "learning_rate": 0.0002,
+      "loss": 1.3825,
+      "step": 4730
+    },
+    {
+      "epoch": 3.678696158323632,
+      "grad_norm": 0.8994188904762268,
+      "learning_rate": 0.0002,
+      "loss": 1.4213,
+      "step": 4740
+    },
+    {
+      "epoch": 3.686457120682965,
+      "grad_norm": 0.75416100025177,
+      "learning_rate": 0.0002,
+      "loss": 1.4281,
+      "step": 4750
+    },
+    {
+      "epoch": 3.6942180830422973,
+      "grad_norm": 0.6499266028404236,
+      "learning_rate": 0.0002,
+      "loss": 1.4154,
+      "step": 4760
+    },
+    {
+      "epoch": 3.7019790454016297,
+      "grad_norm": 0.7246791124343872,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7097400077609626,
+      "grad_norm": 0.7831124067306519,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 4780
+    },
+    {
+      "epoch": 3.717500970120295,
+      "grad_norm": 0.7130028009414673,
+      "learning_rate": 0.0002,
+      "loss": 1.3933,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7252619324796274,
+      "grad_norm": 0.7501602172851562,
+      "learning_rate": 0.0002,
+      "loss": 1.4632,
+      "step": 4800
+    },
+    {
+      "epoch": 3.73302289483896,
+      "grad_norm": 0.6980932950973511,
+      "learning_rate": 0.0002,
+      "loss": 1.4985,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7407838571982923,
+      "grad_norm": 0.8050530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.4517,
+      "step": 4820
+    },
+    {
+      "epoch": 3.748544819557625,
+      "grad_norm": 0.6385579705238342,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7563057819169576,
+      "grad_norm": 0.6664714813232422,
+      "learning_rate": 0.0002,
+      "loss": 1.5281,
+      "step": 4840
+    },
+    {
+      "epoch": 3.76406674427629,
+      "grad_norm": 0.7125676274299622,
+      "learning_rate": 0.0002,
+      "loss": 1.4443,
+      "step": 4850
+    },
+    {
+      "epoch": 3.771827706635623,
+      "grad_norm": 0.7231866717338562,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4860
+    },
+    {
+      "epoch": 3.7795886689949554,
+      "grad_norm": 0.6917183995246887,
+      "learning_rate": 0.0002,
+      "loss": 1.4446,
+      "step": 4870
+    },
+    {
+      "epoch": 3.787349631354288,
+      "grad_norm": 0.665037989616394,
+      "learning_rate": 0.0002,
+      "loss": 1.4369,
+      "step": 4880
+    },
+    {
+      "epoch": 3.7951105937136207,
+      "grad_norm": 0.5837726593017578,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4890
+    },
+    {
+      "epoch": 3.802871556072953,
+      "grad_norm": 0.6366701722145081,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8106325184322856,
+      "grad_norm": 0.7082223892211914,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 4910
+    },
+    {
+      "epoch": 3.818393480791618,
+      "grad_norm": 0.8101672530174255,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4920
+    },
+    {
+      "epoch": 3.826154443150951,
+      "grad_norm": 0.7516148090362549,
+      "learning_rate": 0.0002,
+      "loss": 1.3659,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8339154055102833,
+      "grad_norm": 0.7928489446640015,
+      "learning_rate": 0.0002,
+      "loss": 1.3909,
+      "step": 4940
+    },
+    {
+      "epoch": 3.8416763678696157,
+      "grad_norm": 0.6892234683036804,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 4950
+    },
+    {
+      "epoch": 3.849437330228948,
+      "grad_norm": 0.6381304264068604,
+      "learning_rate": 0.0002,
+      "loss": 1.5024,
+      "step": 4960
+    },
+    {
+      "epoch": 3.857198292588281,
+      "grad_norm": 0.8068831562995911,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8649592549476135,
+      "grad_norm": 0.7289869785308838,
+      "learning_rate": 0.0002,
+      "loss": 1.45,
+      "step": 4980
+    },
+    {
+      "epoch": 3.872720217306946,
+      "grad_norm": 0.7278549075126648,
+      "learning_rate": 0.0002,
+      "loss": 1.398,
+      "step": 4990
+    },
+    {
+      "epoch": 3.880481179666279,
+      "grad_norm": 0.7324236631393433,
+      "learning_rate": 0.0002,
+      "loss": 1.4442,
+      "step": 5000
+    },
+    {
+      "epoch": 3.8882421420256112,
+      "grad_norm": 0.6759871244430542,
+      "learning_rate": 0.0002,
+      "loss": 1.4511,
+      "step": 5010
+    },
+    {
+      "epoch": 3.8960031043849437,
+      "grad_norm": 0.8159207701683044,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 5020
+    },
+    {
+      "epoch": 3.9037640667442766,
+      "grad_norm": 0.6536211967468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5030
+    },
+    {
+      "epoch": 3.911525029103609,
+      "grad_norm": 0.6827932000160217,
+      "learning_rate": 0.0002,
+      "loss": 1.4335,
+      "step": 5040
+    },
+    {
+      "epoch": 3.9192859914629414,
+      "grad_norm": 0.6688340306282043,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 5050
+    },
+    {
+      "epoch": 3.927046953822274,
+      "grad_norm": 0.6385695934295654,
+      "learning_rate": 0.0002,
+      "loss": 1.4099,
+      "step": 5060
+    },
+    {
+      "epoch": 3.9348079161816063,
+      "grad_norm": 0.6975107192993164,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 5070
+    },
+    {
+      "epoch": 3.942568878540939,
+      "grad_norm": 0.6684112548828125,
+      "learning_rate": 0.0002,
+      "loss": 1.4893,
+      "step": 5080
+    },
+    {
+      "epoch": 3.9503298409002716,
+      "grad_norm": 0.8349628448486328,
+      "learning_rate": 0.0002,
+      "loss": 1.4732,
+      "step": 5090
+    },
+    {
+      "epoch": 3.958090803259604,
+      "grad_norm": 0.7146425843238831,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 5100
+    },
+    {
+      "epoch": 3.965851765618937,
+      "grad_norm": 0.6555036902427673,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9736127279782694,
+      "grad_norm": 0.7037415504455566,
+      "learning_rate": 0.0002,
+      "loss": 1.4274,
+      "step": 5120
+    },
+    {
+      "epoch": 3.981373690337602,
+      "grad_norm": 0.7235575914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5130
+    },
+    {
+      "epoch": 3.9891346526969347,
+      "grad_norm": 0.7092325687408447,
+      "learning_rate": 0.0002,
+      "loss": 1.4455,
+      "step": 5140
+    },
+    {
+      "epoch": 3.996895615056267,
+      "grad_norm": 0.7490319609642029,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 5150
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9131355285644531,
+      "eval_runtime": 105.5778,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 5154
+    },
+    {
+      "epoch": 4.0046565774155995,
+      "grad_norm": 0.7075854539871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 5160
+    },
+    {
+      "epoch": 4.012417539774932,
+      "grad_norm": 0.9466007351875305,
+      "learning_rate": 0.0002,
+      "loss": 1.209,
+      "step": 5170
+    },
+    {
+      "epoch": 4.020178502134264,
+      "grad_norm": 1.0297044515609741,
+      "learning_rate": 0.0002,
+      "loss": 1.2567,
+      "step": 5180
+    },
+    {
+      "epoch": 4.027939464493597,
+      "grad_norm": 0.7765059471130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5190
+    },
+    {
+      "epoch": 4.03570042685293,
+      "grad_norm": 0.995760977268219,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 5200
+    },
+    {
+      "epoch": 4.043461389212262,
+      "grad_norm": 0.8663829565048218,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 5210
+    },
+    {
+      "epoch": 4.051222351571595,
+      "grad_norm": 1.0660825967788696,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 5220
+    },
+    {
+      "epoch": 4.058983313930927,
+      "grad_norm": 0.9858174920082092,
+      "learning_rate": 0.0002,
+      "loss": 1.1676,
+      "step": 5230
+    },
+    {
+      "epoch": 4.06674427629026,
+      "grad_norm": 0.8911338448524475,
+      "learning_rate": 0.0002,
+      "loss": 1.2448,
+      "step": 5240
+    },
+    {
+      "epoch": 4.074505238649593,
+      "grad_norm": 1.0848394632339478,
+      "learning_rate": 0.0002,
+      "loss": 1.1858,
+      "step": 5250
+    },
+    {
+      "epoch": 4.082266201008925,
+      "grad_norm": 1.0849905014038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5260
+    },
+    {
+      "epoch": 4.090027163368258,
+      "grad_norm": 1.0497841835021973,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 5270
+    },
+    {
+      "epoch": 4.0977881257275905,
+      "grad_norm": 0.8943053483963013,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 5280
+    },
+    {
+      "epoch": 4.1055490880869225,
+      "grad_norm": 0.8432527184486389,
+      "learning_rate": 0.0002,
+      "loss": 1.1923,
+      "step": 5290
+    },
+    {
+      "epoch": 4.113310050446255,
+      "grad_norm": 0.9690414667129517,
+      "learning_rate": 0.0002,
+      "loss": 1.1634,
+      "step": 5300
+    },
+    {
+      "epoch": 4.121071012805588,
+      "grad_norm": 0.7790773510932922,
+      "learning_rate": 0.0002,
+      "loss": 1.3019,
+      "step": 5310
+    },
+    {
+      "epoch": 4.12883197516492,
+      "grad_norm": 0.9289211630821228,
+      "learning_rate": 0.0002,
+      "loss": 1.1806,
+      "step": 5320
+    },
+    {
+      "epoch": 4.136592937524253,
+      "grad_norm": 1.0785125494003296,
+      "learning_rate": 0.0002,
+      "loss": 1.1458,
+      "step": 5330
+    },
+    {
+      "epoch": 4.144353899883585,
+      "grad_norm": 0.8559591770172119,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5340
+    },
+    {
+      "epoch": 4.152114862242918,
+      "grad_norm": 0.9405956268310547,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5350
+    },
+    {
+      "epoch": 4.159875824602251,
+      "grad_norm": 0.9942827820777893,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 5360
+    },
+    {
+      "epoch": 4.167636786961583,
+      "grad_norm": 0.9141933917999268,
+      "learning_rate": 0.0002,
+      "loss": 1.1659,
+      "step": 5370
+    },
+    {
+      "epoch": 4.175397749320916,
+      "grad_norm": 0.8206015229225159,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 5380
+    },
+    {
+      "epoch": 4.183158711680249,
+      "grad_norm": 0.9340888857841492,
+      "learning_rate": 0.0002,
+      "loss": 1.2778,
+      "step": 5390
+    },
+    {
+      "epoch": 4.190919674039581,
+      "grad_norm": 1.2122114896774292,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 5400
+    },
+    {
+      "epoch": 4.1986806363989135,
+      "grad_norm": 1.0661298036575317,
+      "learning_rate": 0.0002,
+      "loss": 1.2371,
+      "step": 5410
+    },
+    {
+      "epoch": 4.206441598758246,
+      "grad_norm": 0.9372861385345459,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 5420
+    },
+    {
+      "epoch": 4.214202561117578,
+      "grad_norm": 0.894012987613678,
+      "learning_rate": 0.0002,
+      "loss": 1.2653,
+      "step": 5430
+    },
+    {
+      "epoch": 4.221963523476911,
+      "grad_norm": 1.0647753477096558,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5440
+    },
+    {
+      "epoch": 4.229724485836243,
+      "grad_norm": 0.989179790019989,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 5450
+    },
+    {
+      "epoch": 4.237485448195576,
+      "grad_norm": 1.1601181030273438,
+      "learning_rate": 0.0002,
+      "loss": 1.2715,
+      "step": 5460
+    },
+    {
+      "epoch": 4.245246410554909,
+      "grad_norm": 0.9395585656166077,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 5470
+    },
+    {
+      "epoch": 4.253007372914241,
+      "grad_norm": 0.9527766108512878,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 5480
+    },
+    {
+      "epoch": 4.260768335273574,
+      "grad_norm": 1.0319520235061646,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 5490
+    },
+    {
+      "epoch": 4.268529297632907,
+      "grad_norm": 0.8659824728965759,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 5500
+    },
+    {
+      "epoch": 4.276290259992239,
+      "grad_norm": 1.099211573600769,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 5510
+    },
+    {
+      "epoch": 4.284051222351572,
+      "grad_norm": 0.9363361597061157,
+      "learning_rate": 0.0002,
+      "loss": 1.2508,
+      "step": 5520
+    },
+    {
+      "epoch": 4.2918121847109045,
+      "grad_norm": 0.8437647223472595,
+      "learning_rate": 0.0002,
+      "loss": 1.189,
+      "step": 5530
+    },
+    {
+      "epoch": 4.2995731470702365,
+      "grad_norm": 0.9181258678436279,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5540
+    },
+    {
+      "epoch": 4.307334109429569,
+      "grad_norm": 0.9059357643127441,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 5550
+    },
+    {
+      "epoch": 4.315095071788901,
+      "grad_norm": 0.9337241649627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2189,
+      "step": 5560
+    },
+    {
+      "epoch": 4.322856034148234,
+      "grad_norm": 0.9428889155387878,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 5570
+    },
+    {
+      "epoch": 4.330616996507567,
+      "grad_norm": 1.003589153289795,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 5580
+    },
+    {
+      "epoch": 4.338377958866899,
+      "grad_norm": 1.1249268054962158,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 5590
+    },
+    {
+      "epoch": 4.346138921226232,
+      "grad_norm": 0.8623469471931458,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 5600
+    },
+    {
+      "epoch": 4.353899883585565,
+      "grad_norm": 1.1389174461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5610
+    },
+    {
+      "epoch": 4.361660845944897,
+      "grad_norm": 1.0136264562606812,
+      "learning_rate": 0.0002,
+      "loss": 1.2245,
+      "step": 5620
+    },
+    {
+      "epoch": 4.36942180830423,
+      "grad_norm": 0.9567070603370667,
+      "learning_rate": 0.0002,
+      "loss": 1.3473,
+      "step": 5630
+    },
+    {
+      "epoch": 4.377182770663563,
+      "grad_norm": 1.0592148303985596,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 5640
+    },
+    {
+      "epoch": 4.384943733022895,
+      "grad_norm": 1.0110485553741455,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5650
+    },
+    {
+      "epoch": 4.3927046953822275,
+      "grad_norm": 0.9914907217025757,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5660
+    },
+    {
+      "epoch": 4.4004656577415595,
+      "grad_norm": 0.9447247982025146,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 5670
+    },
+    {
+      "epoch": 4.408226620100892,
+      "grad_norm": 0.9644378423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 5680
+    },
+    {
+      "epoch": 4.415987582460225,
+      "grad_norm": 0.920676589012146,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 5690
+    },
+    {
+      "epoch": 4.423748544819557,
+      "grad_norm": 1.060570478439331,
+      "learning_rate": 0.0002,
+      "loss": 1.2792,
+      "step": 5700
+    },
+    {
+      "epoch": 4.43150950717889,
+      "grad_norm": 0.8857738971710205,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5710
+    },
+    {
+      "epoch": 4.439270469538223,
+      "grad_norm": 1.0536398887634277,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 5720
+    },
+    {
+      "epoch": 4.447031431897555,
+      "grad_norm": 0.990847110748291,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 5730
+    },
+    {
+      "epoch": 4.454792394256888,
+      "grad_norm": 0.9692499041557312,
+      "learning_rate": 0.0002,
+      "loss": 1.2469,
+      "step": 5740
+    },
+    {
+      "epoch": 4.462553356616221,
+      "grad_norm": 1.0376402139663696,
+      "learning_rate": 0.0002,
+      "loss": 1.2269,
+      "step": 5750
+    },
+    {
+      "epoch": 4.470314318975553,
+      "grad_norm": 1.3863259553909302,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 5760
+    },
+    {
+      "epoch": 4.478075281334886,
+      "grad_norm": 0.978379487991333,
+      "learning_rate": 0.0002,
+      "loss": 1.2591,
+      "step": 5770
+    },
+    {
+      "epoch": 4.485836243694218,
+      "grad_norm": 1.0973085165023804,
+      "learning_rate": 0.0002,
+      "loss": 1.2729,
+      "step": 5780
+    },
+    {
+      "epoch": 4.4935972060535505,
+      "grad_norm": 1.057006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5790
+    },
+    {
+      "epoch": 4.501358168412883,
+      "grad_norm": 0.9247729182243347,
+      "learning_rate": 0.0002,
+      "loss": 1.2476,
+      "step": 5800
+    },
+    {
+      "epoch": 4.509119130772215,
+      "grad_norm": 1.0447787046432495,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 5810
+    },
+    {
+      "epoch": 4.516880093131548,
+      "grad_norm": 1.1930429935455322,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 5820
+    },
+    {
+      "epoch": 4.524641055490881,
+      "grad_norm": 0.9867590069770813,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5830
+    },
+    {
+      "epoch": 4.532402017850213,
+      "grad_norm": 0.9591100215911865,
+      "learning_rate": 0.0002,
+      "loss": 1.2766,
+      "step": 5840
+    },
+    {
+      "epoch": 4.540162980209546,
+      "grad_norm": 0.9950753450393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 5850
+    },
+    {
+      "epoch": 4.547923942568879,
+      "grad_norm": 1.0087506771087646,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 5860
+    },
+    {
+      "epoch": 4.555684904928211,
+      "grad_norm": 1.0934417247772217,
+      "learning_rate": 0.0002,
+      "loss": 1.3165,
+      "step": 5870
+    },
+    {
+      "epoch": 4.563445867287544,
+      "grad_norm": 1.107987403869629,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 5880
+    },
+    {
+      "epoch": 4.571206829646876,
+      "grad_norm": 0.9147276878356934,
+      "learning_rate": 0.0002,
+      "loss": 1.2184,
+      "step": 5890
+    },
+    {
+      "epoch": 4.578967792006209,
+      "grad_norm": 1.036780595779419,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 5900
+    },
+    {
+      "epoch": 4.5867287543655415,
+      "grad_norm": 0.9284719824790955,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 5910
+    },
+    {
+      "epoch": 4.5944897167248735,
+      "grad_norm": 0.9141898155212402,
+      "learning_rate": 0.0002,
+      "loss": 1.3693,
+      "step": 5920
+    },
+    {
+      "epoch": 4.602250679084206,
+      "grad_norm": 1.0447357892990112,
+      "learning_rate": 0.0002,
+      "loss": 1.2319,
+      "step": 5930
+    },
+    {
+      "epoch": 4.610011641443539,
+      "grad_norm": 0.9309114217758179,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 5940
+    },
+    {
+      "epoch": 4.617772603802871,
+      "grad_norm": 1.2986129522323608,
+      "learning_rate": 0.0002,
+      "loss": 1.2827,
+      "step": 5950
+    },
+    {
+      "epoch": 4.625533566162204,
+      "grad_norm": 0.9221704602241516,
+      "learning_rate": 0.0002,
+      "loss": 1.312,
+      "step": 5960
+    },
+    {
+      "epoch": 4.633294528521537,
+      "grad_norm": 0.9228187799453735,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 5970
+    },
+    {
+      "epoch": 4.641055490880869,
+      "grad_norm": 0.9483116269111633,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 5980
+    },
+    {
+      "epoch": 4.648816453240202,
+      "grad_norm": 1.0218974351882935,
+      "learning_rate": 0.0002,
+      "loss": 1.3437,
+      "step": 5990
+    },
+    {
+      "epoch": 4.656577415599534,
+      "grad_norm": 0.9764600396156311,
+      "learning_rate": 0.0002,
+      "loss": 1.3085,
+      "step": 6000
+    },
+    {
+      "epoch": 4.664338377958867,
+      "grad_norm": 0.9115710258483887,
+      "learning_rate": 0.0002,
+      "loss": 1.197,
+      "step": 6010
+    },
+    {
+      "epoch": 4.6720993403182,
+      "grad_norm": 0.9245651364326477,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 6020
+    },
+    {
+      "epoch": 4.6798603026775325,
+      "grad_norm": 0.9686311483383179,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 6030
+    },
+    {
+      "epoch": 4.6876212650368645,
+      "grad_norm": 1.1807392835617065,
+      "learning_rate": 0.0002,
+      "loss": 1.2702,
+      "step": 6040
+    },
+    {
+      "epoch": 4.695382227396197,
+      "grad_norm": 1.0358641147613525,
+      "learning_rate": 0.0002,
+      "loss": 1.328,
+      "step": 6050
+    },
+    {
+      "epoch": 4.703143189755529,
+      "grad_norm": 0.987332284450531,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6060
+    },
+    {
+      "epoch": 4.710904152114862,
+      "grad_norm": 1.0526494979858398,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6070
+    },
+    {
+      "epoch": 4.718665114474195,
+      "grad_norm": 1.0276758670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.2246,
+      "step": 6080
+    },
+    {
+      "epoch": 4.726426076833527,
+      "grad_norm": 0.9904406666755676,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6090
+    },
+    {
+      "epoch": 4.73418703919286,
+      "grad_norm": 1.0084882974624634,
+      "learning_rate": 0.0002,
+      "loss": 1.2797,
+      "step": 6100
+    },
+    {
+      "epoch": 4.741948001552192,
+      "grad_norm": 0.8646450638771057,
+      "learning_rate": 0.0002,
+      "loss": 1.2656,
+      "step": 6110
+    },
+    {
+      "epoch": 4.749708963911525,
+      "grad_norm": 0.9233377575874329,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 6120
+    },
+    {
+      "epoch": 4.757469926270858,
+      "grad_norm": 0.9675140976905823,
+      "learning_rate": 0.0002,
+      "loss": 1.2642,
+      "step": 6130
+    },
+    {
+      "epoch": 4.765230888630191,
+      "grad_norm": 0.9639796018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6140
+    },
+    {
+      "epoch": 4.772991850989523,
+      "grad_norm": 0.925199568271637,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 6150
+    },
+    {
+      "epoch": 4.7807528133488555,
+      "grad_norm": 1.050901174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 6160
+    },
+    {
+      "epoch": 4.7885137757081875,
+      "grad_norm": 0.8920623660087585,
+      "learning_rate": 0.0002,
+      "loss": 1.301,
+      "step": 6170
+    },
+    {
+      "epoch": 4.79627473806752,
+      "grad_norm": 0.8964757919311523,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6180
+    },
+    {
+      "epoch": 4.804035700426853,
+      "grad_norm": 1.0839070081710815,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 6190
+    },
+    {
+      "epoch": 4.811796662786185,
+      "grad_norm": 0.8809942007064819,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 6200
+    },
+    {
+      "epoch": 4.819557625145518,
+      "grad_norm": 1.0216195583343506,
+      "learning_rate": 0.0002,
+      "loss": 1.321,
+      "step": 6210
+    },
+    {
+      "epoch": 4.827318587504851,
+      "grad_norm": 0.892005980014801,
+      "learning_rate": 0.0002,
+      "loss": 1.3033,
+      "step": 6220
+    },
+    {
+      "epoch": 4.835079549864183,
+      "grad_norm": 0.9957166910171509,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6230
+    },
+    {
+      "epoch": 4.842840512223516,
+      "grad_norm": 0.9720533490180969,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 6240
+    },
+    {
+      "epoch": 4.850601474582849,
+      "grad_norm": 0.9336182475090027,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 6250
+    },
+    {
+      "epoch": 4.858362436942181,
+      "grad_norm": 1.2611457109451294,
+      "learning_rate": 0.0002,
+      "loss": 1.3136,
+      "step": 6260
+    },
+    {
+      "epoch": 4.866123399301514,
+      "grad_norm": 0.8927203416824341,
+      "learning_rate": 0.0002,
+      "loss": 1.2234,
+      "step": 6270
+    },
+    {
+      "epoch": 4.873884361660846,
+      "grad_norm": 0.9706710577011108,
+      "learning_rate": 0.0002,
+      "loss": 1.3463,
+      "step": 6280
+    },
+    {
+      "epoch": 4.8816453240201785,
+      "grad_norm": 1.1461690664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 6290
+    },
+    {
+      "epoch": 4.889406286379511,
+      "grad_norm": 0.9930381178855896,
+      "learning_rate": 0.0002,
+      "loss": 1.2566,
+      "step": 6300
+    },
+    {
+      "epoch": 4.897167248738843,
+      "grad_norm": 0.91451096534729,
+      "learning_rate": 0.0002,
+      "loss": 1.2568,
+      "step": 6310
+    },
+    {
+      "epoch": 4.904928211098176,
+      "grad_norm": 1.0319571495056152,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 6320
+    },
+    {
+      "epoch": 4.912689173457509,
+      "grad_norm": 0.990140438079834,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6330
+    },
+    {
+      "epoch": 4.920450135816841,
+      "grad_norm": 1.2466117143630981,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 6340
+    },
+    {
+      "epoch": 4.928211098176174,
+      "grad_norm": 1.0316979885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 6350
+    },
+    {
+      "epoch": 4.935972060535507,
+      "grad_norm": 1.0643759965896606,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 6360
+    },
+    {
+      "epoch": 4.943733022894839,
+      "grad_norm": 0.9703279733657837,
+      "learning_rate": 0.0002,
+      "loss": 1.2559,
+      "step": 6370
+    },
+    {
+      "epoch": 4.951493985254172,
+      "grad_norm": 0.9767927527427673,
+      "learning_rate": 0.0002,
+      "loss": 1.2155,
+      "step": 6380
+    },
+    {
+      "epoch": 4.959254947613504,
+      "grad_norm": 0.960854172706604,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 6390
+    },
+    {
+      "epoch": 4.967015909972837,
+      "grad_norm": 0.9922910332679749,
+      "learning_rate": 0.0002,
+      "loss": 1.3314,
+      "step": 6400
+    },
+    {
+      "epoch": 4.9747768723321695,
+      "grad_norm": 0.956470787525177,
+      "learning_rate": 0.0002,
+      "loss": 1.3018,
+      "step": 6410
+    },
+    {
+      "epoch": 4.9825378346915015,
+      "grad_norm": 0.9637242555618286,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 6420
+    },
+    {
+      "epoch": 4.990298797050834,
+      "grad_norm": 1.0855202674865723,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 6430
+    },
+    {
+      "epoch": 4.998059759410167,
+      "grad_norm": 0.9655316472053528,
+      "learning_rate": 0.0002,
+      "loss": 1.3015,
+      "step": 6440
+    },
+    {
+      "epoch": 4.9996119518820334,
+      "eval_loss": 2.0410802364349365,
+      "eval_runtime": 113.04,
+      "eval_samples_per_second": 4.485,
+      "eval_steps_per_second": 0.566,
+      "step": 6442
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 3.309109312094208e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-6442/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..581ba80957e863c91c09289d332dfb5bd46a2cf2
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79a559981a68c2cfc898167d3bfe19ceee9c9b2fa121024a67535beb73418906
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9e82a3732717d575e777685b34150335617cf6e7
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7abbfd795613bb333de7a6ae2fcb4c78425f14d5e88acafbdbd037807c472e40
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d0af753f1792b5cc65461c7784b669deedf3fe78
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b77f041a06d5be44bc9d0380f05f012100f3636dcdba5d24755612144d65a1a
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ceb65a5bd14c886ea05ef379f95cb3f922456a34
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2547863cfabcec9c6cd9f2b203f3c8a6cc62dc0289e47e5524ad0ba794730015
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9796c0359b3c23156ed00fa8a9325539c4a7ebec
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/trainer_state.json
@@ -0,0 +1,5492 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 7731,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    },
+    {
+      "epoch": 3.0034924330616994,
+      "grad_norm": 0.5645653605461121,
+      "learning_rate": 0.0002,
+      "loss": 1.4983,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0112533954210323,
+      "grad_norm": 0.6457151174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0190143577803648,
+      "grad_norm": 0.583838164806366,
+      "learning_rate": 0.0002,
+      "loss": 1.3899,
+      "step": 3890
+    },
+    {
+      "epoch": 3.026775320139697,
+      "grad_norm": 0.6819260120391846,
+      "learning_rate": 0.0002,
+      "loss": 1.3258,
+      "step": 3900
+    },
+    {
+      "epoch": 3.03453628249903,
+      "grad_norm": 0.6692903637886047,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0422972448583625,
+      "grad_norm": 0.6101024746894836,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 3920
+    },
+    {
+      "epoch": 3.050058207217695,
+      "grad_norm": 0.7014093399047852,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3930
+    },
+    {
+      "epoch": 3.0578191695770274,
+      "grad_norm": 0.7380381226539612,
+      "learning_rate": 0.0002,
+      "loss": 1.3885,
+      "step": 3940
+    },
+    {
+      "epoch": 3.0655801319363603,
+      "grad_norm": 0.6607900857925415,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0733410942956927,
+      "grad_norm": 0.735263466835022,
+      "learning_rate": 0.0002,
+      "loss": 1.4293,
+      "step": 3960
+    },
+    {
+      "epoch": 3.081102056655025,
+      "grad_norm": 0.6788513660430908,
+      "learning_rate": 0.0002,
+      "loss": 1.3966,
+      "step": 3970
+    },
+    {
+      "epoch": 3.088863019014358,
+      "grad_norm": 0.6347652673721313,
+      "learning_rate": 0.0002,
+      "loss": 1.3435,
+      "step": 3980
+    },
+    {
+      "epoch": 3.0966239813736904,
+      "grad_norm": 0.7056642770767212,
+      "learning_rate": 0.0002,
+      "loss": 1.4518,
+      "step": 3990
+    },
+    {
+      "epoch": 3.104384943733023,
+      "grad_norm": 0.6387075185775757,
+      "learning_rate": 0.0002,
+      "loss": 1.4474,
+      "step": 4000
+    },
+    {
+      "epoch": 3.1121459060923553,
+      "grad_norm": 0.6701116561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.3833,
+      "step": 4010
+    },
+    {
+      "epoch": 3.119906868451688,
+      "grad_norm": 0.7558449506759644,
+      "learning_rate": 0.0002,
+      "loss": 1.404,
+      "step": 4020
+    },
+    {
+      "epoch": 3.1276678308110206,
+      "grad_norm": 0.6612881422042847,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 4030
+    },
+    {
+      "epoch": 3.135428793170353,
+      "grad_norm": 0.7474587559700012,
+      "learning_rate": 0.0002,
+      "loss": 1.439,
+      "step": 4040
+    },
+    {
+      "epoch": 3.1431897555296855,
+      "grad_norm": 0.7292373776435852,
+      "learning_rate": 0.0002,
+      "loss": 1.4616,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1509507178890184,
+      "grad_norm": 0.7432886958122253,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 4060
+    },
+    {
+      "epoch": 3.158711680248351,
+      "grad_norm": 0.6366098523139954,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1664726426076832,
+      "grad_norm": 0.6837611794471741,
+      "learning_rate": 0.0002,
+      "loss": 1.5044,
+      "step": 4080
+    },
+    {
+      "epoch": 3.174233604967016,
+      "grad_norm": 0.7194393277168274,
+      "learning_rate": 0.0002,
+      "loss": 1.4332,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1819945673263486,
+      "grad_norm": 0.6963607668876648,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 4100
+    },
+    {
+      "epoch": 3.189755529685681,
+      "grad_norm": 0.6404902935028076,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4110
+    },
+    {
+      "epoch": 3.1975164920450134,
+      "grad_norm": 0.7172070741653442,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 4120
+    },
+    {
+      "epoch": 3.2052774544043463,
+      "grad_norm": 0.6577759385108948,
+      "learning_rate": 0.0002,
+      "loss": 1.4658,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2130384167636787,
+      "grad_norm": 0.6658480167388916,
+      "learning_rate": 0.0002,
+      "loss": 1.4019,
+      "step": 4140
+    },
+    {
+      "epoch": 3.220799379123011,
+      "grad_norm": 0.6771699786186218,
+      "learning_rate": 0.0002,
+      "loss": 1.4348,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2285603414823436,
+      "grad_norm": 0.699035108089447,
+      "learning_rate": 0.0002,
+      "loss": 1.4736,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2363213038416765,
+      "grad_norm": 0.7218514680862427,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 4170
+    },
+    {
+      "epoch": 3.244082266201009,
+      "grad_norm": 0.6270631551742554,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 4180
+    },
+    {
+      "epoch": 3.2518432285603414,
+      "grad_norm": 0.6828921437263489,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2596041909196742,
+      "grad_norm": 0.6005498170852661,
+      "learning_rate": 0.0002,
+      "loss": 1.4663,
+      "step": 4200
+    },
+    {
+      "epoch": 3.2673651532790067,
+      "grad_norm": 0.6974790692329407,
+      "learning_rate": 0.0002,
+      "loss": 1.4798,
+      "step": 4210
+    },
+    {
+      "epoch": 3.275126115638339,
+      "grad_norm": 0.7269543409347534,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4220
+    },
+    {
+      "epoch": 3.2828870779976715,
+      "grad_norm": 0.6728787422180176,
+      "learning_rate": 0.0002,
+      "loss": 1.3848,
+      "step": 4230
+    },
+    {
+      "epoch": 3.2906480403570044,
+      "grad_norm": 0.676972508430481,
+      "learning_rate": 0.0002,
+      "loss": 1.4112,
+      "step": 4240
+    },
+    {
+      "epoch": 3.298409002716337,
+      "grad_norm": 0.748309314250946,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3061699650756693,
+      "grad_norm": 0.6976589560508728,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 4260
+    },
+    {
+      "epoch": 3.3139309274350017,
+      "grad_norm": 0.649780809879303,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3216918897943346,
+      "grad_norm": 0.6529902815818787,
+      "learning_rate": 0.0002,
+      "loss": 1.327,
+      "step": 4280
+    },
+    {
+      "epoch": 3.329452852153667,
+      "grad_norm": 0.9273163676261902,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 4290
+    },
+    {
+      "epoch": 3.3372138145129995,
+      "grad_norm": 0.717024028301239,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 4300
+    },
+    {
+      "epoch": 3.3449747768723324,
+      "grad_norm": 0.7914950251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.4441,
+      "step": 4310
+    },
+    {
+      "epoch": 3.352735739231665,
+      "grad_norm": 0.7133203148841858,
+      "learning_rate": 0.0002,
+      "loss": 1.432,
+      "step": 4320
+    },
+    {
+      "epoch": 3.3604967015909972,
+      "grad_norm": 0.7409568428993225,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3682576639503297,
+      "grad_norm": 0.6993981003761292,
+      "learning_rate": 0.0002,
+      "loss": 1.3992,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3760186263096625,
+      "grad_norm": 0.7114535570144653,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4350
+    },
+    {
+      "epoch": 3.383779588668995,
+      "grad_norm": 0.6790860295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.4227,
+      "step": 4360
+    },
+    {
+      "epoch": 3.3915405510283274,
+      "grad_norm": 0.6507849097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 4370
+    },
+    {
+      "epoch": 3.39930151338766,
+      "grad_norm": 0.5967804193496704,
+      "learning_rate": 0.0002,
+      "loss": 1.4559,
+      "step": 4380
+    },
+    {
+      "epoch": 3.4070624757469927,
+      "grad_norm": 0.6625847816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 4390
+    },
+    {
+      "epoch": 3.414823438106325,
+      "grad_norm": 0.6736508011817932,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4225844004656576,
+      "grad_norm": 0.7870860695838928,
+      "learning_rate": 0.0002,
+      "loss": 1.4363,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4303453628249905,
+      "grad_norm": 0.7205295562744141,
+      "learning_rate": 0.0002,
+      "loss": 1.4114,
+      "step": 4420
+    },
+    {
+      "epoch": 3.438106325184323,
+      "grad_norm": 0.6634634137153625,
+      "learning_rate": 0.0002,
+      "loss": 1.4131,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4458672875436553,
+      "grad_norm": 0.7562733292579651,
+      "learning_rate": 0.0002,
+      "loss": 1.4683,
+      "step": 4440
+    },
+    {
+      "epoch": 3.453628249902988,
+      "grad_norm": 0.6585879921913147,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4613892122623207,
+      "grad_norm": 0.6896792054176331,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 4460
+    },
+    {
+      "epoch": 3.469150174621653,
+      "grad_norm": 0.6520342230796814,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4769111369809855,
+      "grad_norm": 0.6760806441307068,
+      "learning_rate": 0.0002,
+      "loss": 1.3423,
+      "step": 4480
+    },
+    {
+      "epoch": 3.484672099340318,
+      "grad_norm": 0.7539774179458618,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 4490
+    },
+    {
+      "epoch": 3.492433061699651,
+      "grad_norm": 0.7409411668777466,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 4500
+    },
+    {
+      "epoch": 3.5001940240589833,
+      "grad_norm": 0.6876253485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.4069,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5079549864183157,
+      "grad_norm": 0.7028461694717407,
+      "learning_rate": 0.0002,
+      "loss": 1.4228,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5157159487776486,
+      "grad_norm": 0.8056529760360718,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4530
+    },
+    {
+      "epoch": 3.523476911136981,
+      "grad_norm": 0.711338996887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5312378734963135,
+      "grad_norm": 0.7343552708625793,
+      "learning_rate": 0.0002,
+      "loss": 1.5247,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5389988358556463,
+      "grad_norm": 0.745479941368103,
+      "learning_rate": 0.0002,
+      "loss": 1.4308,
+      "step": 4560
+    },
+    {
+      "epoch": 3.5467597982149788,
+      "grad_norm": 0.7582294940948486,
+      "learning_rate": 0.0002,
+      "loss": 1.4229,
+      "step": 4570
+    },
+    {
+      "epoch": 3.554520760574311,
+      "grad_norm": 0.6717444658279419,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4580
+    },
+    {
+      "epoch": 3.5622817229336436,
+      "grad_norm": 0.7417883276939392,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 4590
+    },
+    {
+      "epoch": 3.570042685292976,
+      "grad_norm": 0.6385737061500549,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4600
+    },
+    {
+      "epoch": 3.577803647652309,
+      "grad_norm": 0.716704249382019,
+      "learning_rate": 0.0002,
+      "loss": 1.3981,
+      "step": 4610
+    },
+    {
+      "epoch": 3.5855646100116414,
+      "grad_norm": 0.6948980093002319,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 4620
+    },
+    {
+      "epoch": 3.593325572370974,
+      "grad_norm": 0.6961140036582947,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6010865347303067,
+      "grad_norm": 0.7493122220039368,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 4640
+    },
+    {
+      "epoch": 3.608847497089639,
+      "grad_norm": 0.7431658506393433,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6166084594489716,
+      "grad_norm": 0.8353387713432312,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4660
+    },
+    {
+      "epoch": 3.6243694218083045,
+      "grad_norm": 0.7095612287521362,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 4670
+    },
+    {
+      "epoch": 3.632130384167637,
+      "grad_norm": 0.776620090007782,
+      "learning_rate": 0.0002,
+      "loss": 1.4003,
+      "step": 4680
+    },
+    {
+      "epoch": 3.6398913465269693,
+      "grad_norm": 0.7198925018310547,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6476523088863018,
+      "grad_norm": 0.8238834738731384,
+      "learning_rate": 0.0002,
+      "loss": 1.4543,
+      "step": 4700
+    },
+    {
+      "epoch": 3.655413271245634,
+      "grad_norm": 0.6804245710372925,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4710
+    },
+    {
+      "epoch": 3.663174233604967,
+      "grad_norm": 0.8444845676422119,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 4720
+    },
+    {
+      "epoch": 3.6709351959642995,
+      "grad_norm": 0.743797779083252,
+      "learning_rate": 0.0002,
+      "loss": 1.3825,
+      "step": 4730
+    },
+    {
+      "epoch": 3.678696158323632,
+      "grad_norm": 0.8994188904762268,
+      "learning_rate": 0.0002,
+      "loss": 1.4213,
+      "step": 4740
+    },
+    {
+      "epoch": 3.686457120682965,
+      "grad_norm": 0.75416100025177,
+      "learning_rate": 0.0002,
+      "loss": 1.4281,
+      "step": 4750
+    },
+    {
+      "epoch": 3.6942180830422973,
+      "grad_norm": 0.6499266028404236,
+      "learning_rate": 0.0002,
+      "loss": 1.4154,
+      "step": 4760
+    },
+    {
+      "epoch": 3.7019790454016297,
+      "grad_norm": 0.7246791124343872,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7097400077609626,
+      "grad_norm": 0.7831124067306519,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 4780
+    },
+    {
+      "epoch": 3.717500970120295,
+      "grad_norm": 0.7130028009414673,
+      "learning_rate": 0.0002,
+      "loss": 1.3933,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7252619324796274,
+      "grad_norm": 0.7501602172851562,
+      "learning_rate": 0.0002,
+      "loss": 1.4632,
+      "step": 4800
+    },
+    {
+      "epoch": 3.73302289483896,
+      "grad_norm": 0.6980932950973511,
+      "learning_rate": 0.0002,
+      "loss": 1.4985,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7407838571982923,
+      "grad_norm": 0.8050530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.4517,
+      "step": 4820
+    },
+    {
+      "epoch": 3.748544819557625,
+      "grad_norm": 0.6385579705238342,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7563057819169576,
+      "grad_norm": 0.6664714813232422,
+      "learning_rate": 0.0002,
+      "loss": 1.5281,
+      "step": 4840
+    },
+    {
+      "epoch": 3.76406674427629,
+      "grad_norm": 0.7125676274299622,
+      "learning_rate": 0.0002,
+      "loss": 1.4443,
+      "step": 4850
+    },
+    {
+      "epoch": 3.771827706635623,
+      "grad_norm": 0.7231866717338562,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4860
+    },
+    {
+      "epoch": 3.7795886689949554,
+      "grad_norm": 0.6917183995246887,
+      "learning_rate": 0.0002,
+      "loss": 1.4446,
+      "step": 4870
+    },
+    {
+      "epoch": 3.787349631354288,
+      "grad_norm": 0.665037989616394,
+      "learning_rate": 0.0002,
+      "loss": 1.4369,
+      "step": 4880
+    },
+    {
+      "epoch": 3.7951105937136207,
+      "grad_norm": 0.5837726593017578,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4890
+    },
+    {
+      "epoch": 3.802871556072953,
+      "grad_norm": 0.6366701722145081,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8106325184322856,
+      "grad_norm": 0.7082223892211914,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 4910
+    },
+    {
+      "epoch": 3.818393480791618,
+      "grad_norm": 0.8101672530174255,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4920
+    },
+    {
+      "epoch": 3.826154443150951,
+      "grad_norm": 0.7516148090362549,
+      "learning_rate": 0.0002,
+      "loss": 1.3659,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8339154055102833,
+      "grad_norm": 0.7928489446640015,
+      "learning_rate": 0.0002,
+      "loss": 1.3909,
+      "step": 4940
+    },
+    {
+      "epoch": 3.8416763678696157,
+      "grad_norm": 0.6892234683036804,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 4950
+    },
+    {
+      "epoch": 3.849437330228948,
+      "grad_norm": 0.6381304264068604,
+      "learning_rate": 0.0002,
+      "loss": 1.5024,
+      "step": 4960
+    },
+    {
+      "epoch": 3.857198292588281,
+      "grad_norm": 0.8068831562995911,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8649592549476135,
+      "grad_norm": 0.7289869785308838,
+      "learning_rate": 0.0002,
+      "loss": 1.45,
+      "step": 4980
+    },
+    {
+      "epoch": 3.872720217306946,
+      "grad_norm": 0.7278549075126648,
+      "learning_rate": 0.0002,
+      "loss": 1.398,
+      "step": 4990
+    },
+    {
+      "epoch": 3.880481179666279,
+      "grad_norm": 0.7324236631393433,
+      "learning_rate": 0.0002,
+      "loss": 1.4442,
+      "step": 5000
+    },
+    {
+      "epoch": 3.8882421420256112,
+      "grad_norm": 0.6759871244430542,
+      "learning_rate": 0.0002,
+      "loss": 1.4511,
+      "step": 5010
+    },
+    {
+      "epoch": 3.8960031043849437,
+      "grad_norm": 0.8159207701683044,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 5020
+    },
+    {
+      "epoch": 3.9037640667442766,
+      "grad_norm": 0.6536211967468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5030
+    },
+    {
+      "epoch": 3.911525029103609,
+      "grad_norm": 0.6827932000160217,
+      "learning_rate": 0.0002,
+      "loss": 1.4335,
+      "step": 5040
+    },
+    {
+      "epoch": 3.9192859914629414,
+      "grad_norm": 0.6688340306282043,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 5050
+    },
+    {
+      "epoch": 3.927046953822274,
+      "grad_norm": 0.6385695934295654,
+      "learning_rate": 0.0002,
+      "loss": 1.4099,
+      "step": 5060
+    },
+    {
+      "epoch": 3.9348079161816063,
+      "grad_norm": 0.6975107192993164,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 5070
+    },
+    {
+      "epoch": 3.942568878540939,
+      "grad_norm": 0.6684112548828125,
+      "learning_rate": 0.0002,
+      "loss": 1.4893,
+      "step": 5080
+    },
+    {
+      "epoch": 3.9503298409002716,
+      "grad_norm": 0.8349628448486328,
+      "learning_rate": 0.0002,
+      "loss": 1.4732,
+      "step": 5090
+    },
+    {
+      "epoch": 3.958090803259604,
+      "grad_norm": 0.7146425843238831,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 5100
+    },
+    {
+      "epoch": 3.965851765618937,
+      "grad_norm": 0.6555036902427673,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9736127279782694,
+      "grad_norm": 0.7037415504455566,
+      "learning_rate": 0.0002,
+      "loss": 1.4274,
+      "step": 5120
+    },
+    {
+      "epoch": 3.981373690337602,
+      "grad_norm": 0.7235575914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5130
+    },
+    {
+      "epoch": 3.9891346526969347,
+      "grad_norm": 0.7092325687408447,
+      "learning_rate": 0.0002,
+      "loss": 1.4455,
+      "step": 5140
+    },
+    {
+      "epoch": 3.996895615056267,
+      "grad_norm": 0.7490319609642029,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 5150
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9131355285644531,
+      "eval_runtime": 105.5778,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 5154
+    },
+    {
+      "epoch": 4.0046565774155995,
+      "grad_norm": 0.7075854539871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 5160
+    },
+    {
+      "epoch": 4.012417539774932,
+      "grad_norm": 0.9466007351875305,
+      "learning_rate": 0.0002,
+      "loss": 1.209,
+      "step": 5170
+    },
+    {
+      "epoch": 4.020178502134264,
+      "grad_norm": 1.0297044515609741,
+      "learning_rate": 0.0002,
+      "loss": 1.2567,
+      "step": 5180
+    },
+    {
+      "epoch": 4.027939464493597,
+      "grad_norm": 0.7765059471130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5190
+    },
+    {
+      "epoch": 4.03570042685293,
+      "grad_norm": 0.995760977268219,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 5200
+    },
+    {
+      "epoch": 4.043461389212262,
+      "grad_norm": 0.8663829565048218,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 5210
+    },
+    {
+      "epoch": 4.051222351571595,
+      "grad_norm": 1.0660825967788696,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 5220
+    },
+    {
+      "epoch": 4.058983313930927,
+      "grad_norm": 0.9858174920082092,
+      "learning_rate": 0.0002,
+      "loss": 1.1676,
+      "step": 5230
+    },
+    {
+      "epoch": 4.06674427629026,
+      "grad_norm": 0.8911338448524475,
+      "learning_rate": 0.0002,
+      "loss": 1.2448,
+      "step": 5240
+    },
+    {
+      "epoch": 4.074505238649593,
+      "grad_norm": 1.0848394632339478,
+      "learning_rate": 0.0002,
+      "loss": 1.1858,
+      "step": 5250
+    },
+    {
+      "epoch": 4.082266201008925,
+      "grad_norm": 1.0849905014038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5260
+    },
+    {
+      "epoch": 4.090027163368258,
+      "grad_norm": 1.0497841835021973,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 5270
+    },
+    {
+      "epoch": 4.0977881257275905,
+      "grad_norm": 0.8943053483963013,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 5280
+    },
+    {
+      "epoch": 4.1055490880869225,
+      "grad_norm": 0.8432527184486389,
+      "learning_rate": 0.0002,
+      "loss": 1.1923,
+      "step": 5290
+    },
+    {
+      "epoch": 4.113310050446255,
+      "grad_norm": 0.9690414667129517,
+      "learning_rate": 0.0002,
+      "loss": 1.1634,
+      "step": 5300
+    },
+    {
+      "epoch": 4.121071012805588,
+      "grad_norm": 0.7790773510932922,
+      "learning_rate": 0.0002,
+      "loss": 1.3019,
+      "step": 5310
+    },
+    {
+      "epoch": 4.12883197516492,
+      "grad_norm": 0.9289211630821228,
+      "learning_rate": 0.0002,
+      "loss": 1.1806,
+      "step": 5320
+    },
+    {
+      "epoch": 4.136592937524253,
+      "grad_norm": 1.0785125494003296,
+      "learning_rate": 0.0002,
+      "loss": 1.1458,
+      "step": 5330
+    },
+    {
+      "epoch": 4.144353899883585,
+      "grad_norm": 0.8559591770172119,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5340
+    },
+    {
+      "epoch": 4.152114862242918,
+      "grad_norm": 0.9405956268310547,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5350
+    },
+    {
+      "epoch": 4.159875824602251,
+      "grad_norm": 0.9942827820777893,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 5360
+    },
+    {
+      "epoch": 4.167636786961583,
+      "grad_norm": 0.9141933917999268,
+      "learning_rate": 0.0002,
+      "loss": 1.1659,
+      "step": 5370
+    },
+    {
+      "epoch": 4.175397749320916,
+      "grad_norm": 0.8206015229225159,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 5380
+    },
+    {
+      "epoch": 4.183158711680249,
+      "grad_norm": 0.9340888857841492,
+      "learning_rate": 0.0002,
+      "loss": 1.2778,
+      "step": 5390
+    },
+    {
+      "epoch": 4.190919674039581,
+      "grad_norm": 1.2122114896774292,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 5400
+    },
+    {
+      "epoch": 4.1986806363989135,
+      "grad_norm": 1.0661298036575317,
+      "learning_rate": 0.0002,
+      "loss": 1.2371,
+      "step": 5410
+    },
+    {
+      "epoch": 4.206441598758246,
+      "grad_norm": 0.9372861385345459,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 5420
+    },
+    {
+      "epoch": 4.214202561117578,
+      "grad_norm": 0.894012987613678,
+      "learning_rate": 0.0002,
+      "loss": 1.2653,
+      "step": 5430
+    },
+    {
+      "epoch": 4.221963523476911,
+      "grad_norm": 1.0647753477096558,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5440
+    },
+    {
+      "epoch": 4.229724485836243,
+      "grad_norm": 0.989179790019989,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 5450
+    },
+    {
+      "epoch": 4.237485448195576,
+      "grad_norm": 1.1601181030273438,
+      "learning_rate": 0.0002,
+      "loss": 1.2715,
+      "step": 5460
+    },
+    {
+      "epoch": 4.245246410554909,
+      "grad_norm": 0.9395585656166077,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 5470
+    },
+    {
+      "epoch": 4.253007372914241,
+      "grad_norm": 0.9527766108512878,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 5480
+    },
+    {
+      "epoch": 4.260768335273574,
+      "grad_norm": 1.0319520235061646,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 5490
+    },
+    {
+      "epoch": 4.268529297632907,
+      "grad_norm": 0.8659824728965759,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 5500
+    },
+    {
+      "epoch": 4.276290259992239,
+      "grad_norm": 1.099211573600769,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 5510
+    },
+    {
+      "epoch": 4.284051222351572,
+      "grad_norm": 0.9363361597061157,
+      "learning_rate": 0.0002,
+      "loss": 1.2508,
+      "step": 5520
+    },
+    {
+      "epoch": 4.2918121847109045,
+      "grad_norm": 0.8437647223472595,
+      "learning_rate": 0.0002,
+      "loss": 1.189,
+      "step": 5530
+    },
+    {
+      "epoch": 4.2995731470702365,
+      "grad_norm": 0.9181258678436279,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5540
+    },
+    {
+      "epoch": 4.307334109429569,
+      "grad_norm": 0.9059357643127441,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 5550
+    },
+    {
+      "epoch": 4.315095071788901,
+      "grad_norm": 0.9337241649627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2189,
+      "step": 5560
+    },
+    {
+      "epoch": 4.322856034148234,
+      "grad_norm": 0.9428889155387878,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 5570
+    },
+    {
+      "epoch": 4.330616996507567,
+      "grad_norm": 1.003589153289795,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 5580
+    },
+    {
+      "epoch": 4.338377958866899,
+      "grad_norm": 1.1249268054962158,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 5590
+    },
+    {
+      "epoch": 4.346138921226232,
+      "grad_norm": 0.8623469471931458,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 5600
+    },
+    {
+      "epoch": 4.353899883585565,
+      "grad_norm": 1.1389174461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5610
+    },
+    {
+      "epoch": 4.361660845944897,
+      "grad_norm": 1.0136264562606812,
+      "learning_rate": 0.0002,
+      "loss": 1.2245,
+      "step": 5620
+    },
+    {
+      "epoch": 4.36942180830423,
+      "grad_norm": 0.9567070603370667,
+      "learning_rate": 0.0002,
+      "loss": 1.3473,
+      "step": 5630
+    },
+    {
+      "epoch": 4.377182770663563,
+      "grad_norm": 1.0592148303985596,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 5640
+    },
+    {
+      "epoch": 4.384943733022895,
+      "grad_norm": 1.0110485553741455,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5650
+    },
+    {
+      "epoch": 4.3927046953822275,
+      "grad_norm": 0.9914907217025757,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5660
+    },
+    {
+      "epoch": 4.4004656577415595,
+      "grad_norm": 0.9447247982025146,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 5670
+    },
+    {
+      "epoch": 4.408226620100892,
+      "grad_norm": 0.9644378423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 5680
+    },
+    {
+      "epoch": 4.415987582460225,
+      "grad_norm": 0.920676589012146,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 5690
+    },
+    {
+      "epoch": 4.423748544819557,
+      "grad_norm": 1.060570478439331,
+      "learning_rate": 0.0002,
+      "loss": 1.2792,
+      "step": 5700
+    },
+    {
+      "epoch": 4.43150950717889,
+      "grad_norm": 0.8857738971710205,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5710
+    },
+    {
+      "epoch": 4.439270469538223,
+      "grad_norm": 1.0536398887634277,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 5720
+    },
+    {
+      "epoch": 4.447031431897555,
+      "grad_norm": 0.990847110748291,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 5730
+    },
+    {
+      "epoch": 4.454792394256888,
+      "grad_norm": 0.9692499041557312,
+      "learning_rate": 0.0002,
+      "loss": 1.2469,
+      "step": 5740
+    },
+    {
+      "epoch": 4.462553356616221,
+      "grad_norm": 1.0376402139663696,
+      "learning_rate": 0.0002,
+      "loss": 1.2269,
+      "step": 5750
+    },
+    {
+      "epoch": 4.470314318975553,
+      "grad_norm": 1.3863259553909302,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 5760
+    },
+    {
+      "epoch": 4.478075281334886,
+      "grad_norm": 0.978379487991333,
+      "learning_rate": 0.0002,
+      "loss": 1.2591,
+      "step": 5770
+    },
+    {
+      "epoch": 4.485836243694218,
+      "grad_norm": 1.0973085165023804,
+      "learning_rate": 0.0002,
+      "loss": 1.2729,
+      "step": 5780
+    },
+    {
+      "epoch": 4.4935972060535505,
+      "grad_norm": 1.057006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5790
+    },
+    {
+      "epoch": 4.501358168412883,
+      "grad_norm": 0.9247729182243347,
+      "learning_rate": 0.0002,
+      "loss": 1.2476,
+      "step": 5800
+    },
+    {
+      "epoch": 4.509119130772215,
+      "grad_norm": 1.0447787046432495,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 5810
+    },
+    {
+      "epoch": 4.516880093131548,
+      "grad_norm": 1.1930429935455322,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 5820
+    },
+    {
+      "epoch": 4.524641055490881,
+      "grad_norm": 0.9867590069770813,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5830
+    },
+    {
+      "epoch": 4.532402017850213,
+      "grad_norm": 0.9591100215911865,
+      "learning_rate": 0.0002,
+      "loss": 1.2766,
+      "step": 5840
+    },
+    {
+      "epoch": 4.540162980209546,
+      "grad_norm": 0.9950753450393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 5850
+    },
+    {
+      "epoch": 4.547923942568879,
+      "grad_norm": 1.0087506771087646,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 5860
+    },
+    {
+      "epoch": 4.555684904928211,
+      "grad_norm": 1.0934417247772217,
+      "learning_rate": 0.0002,
+      "loss": 1.3165,
+      "step": 5870
+    },
+    {
+      "epoch": 4.563445867287544,
+      "grad_norm": 1.107987403869629,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 5880
+    },
+    {
+      "epoch": 4.571206829646876,
+      "grad_norm": 0.9147276878356934,
+      "learning_rate": 0.0002,
+      "loss": 1.2184,
+      "step": 5890
+    },
+    {
+      "epoch": 4.578967792006209,
+      "grad_norm": 1.036780595779419,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 5900
+    },
+    {
+      "epoch": 4.5867287543655415,
+      "grad_norm": 0.9284719824790955,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 5910
+    },
+    {
+      "epoch": 4.5944897167248735,
+      "grad_norm": 0.9141898155212402,
+      "learning_rate": 0.0002,
+      "loss": 1.3693,
+      "step": 5920
+    },
+    {
+      "epoch": 4.602250679084206,
+      "grad_norm": 1.0447357892990112,
+      "learning_rate": 0.0002,
+      "loss": 1.2319,
+      "step": 5930
+    },
+    {
+      "epoch": 4.610011641443539,
+      "grad_norm": 0.9309114217758179,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 5940
+    },
+    {
+      "epoch": 4.617772603802871,
+      "grad_norm": 1.2986129522323608,
+      "learning_rate": 0.0002,
+      "loss": 1.2827,
+      "step": 5950
+    },
+    {
+      "epoch": 4.625533566162204,
+      "grad_norm": 0.9221704602241516,
+      "learning_rate": 0.0002,
+      "loss": 1.312,
+      "step": 5960
+    },
+    {
+      "epoch": 4.633294528521537,
+      "grad_norm": 0.9228187799453735,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 5970
+    },
+    {
+      "epoch": 4.641055490880869,
+      "grad_norm": 0.9483116269111633,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 5980
+    },
+    {
+      "epoch": 4.648816453240202,
+      "grad_norm": 1.0218974351882935,
+      "learning_rate": 0.0002,
+      "loss": 1.3437,
+      "step": 5990
+    },
+    {
+      "epoch": 4.656577415599534,
+      "grad_norm": 0.9764600396156311,
+      "learning_rate": 0.0002,
+      "loss": 1.3085,
+      "step": 6000
+    },
+    {
+      "epoch": 4.664338377958867,
+      "grad_norm": 0.9115710258483887,
+      "learning_rate": 0.0002,
+      "loss": 1.197,
+      "step": 6010
+    },
+    {
+      "epoch": 4.6720993403182,
+      "grad_norm": 0.9245651364326477,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 6020
+    },
+    {
+      "epoch": 4.6798603026775325,
+      "grad_norm": 0.9686311483383179,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 6030
+    },
+    {
+      "epoch": 4.6876212650368645,
+      "grad_norm": 1.1807392835617065,
+      "learning_rate": 0.0002,
+      "loss": 1.2702,
+      "step": 6040
+    },
+    {
+      "epoch": 4.695382227396197,
+      "grad_norm": 1.0358641147613525,
+      "learning_rate": 0.0002,
+      "loss": 1.328,
+      "step": 6050
+    },
+    {
+      "epoch": 4.703143189755529,
+      "grad_norm": 0.987332284450531,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6060
+    },
+    {
+      "epoch": 4.710904152114862,
+      "grad_norm": 1.0526494979858398,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6070
+    },
+    {
+      "epoch": 4.718665114474195,
+      "grad_norm": 1.0276758670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.2246,
+      "step": 6080
+    },
+    {
+      "epoch": 4.726426076833527,
+      "grad_norm": 0.9904406666755676,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6090
+    },
+    {
+      "epoch": 4.73418703919286,
+      "grad_norm": 1.0084882974624634,
+      "learning_rate": 0.0002,
+      "loss": 1.2797,
+      "step": 6100
+    },
+    {
+      "epoch": 4.741948001552192,
+      "grad_norm": 0.8646450638771057,
+      "learning_rate": 0.0002,
+      "loss": 1.2656,
+      "step": 6110
+    },
+    {
+      "epoch": 4.749708963911525,
+      "grad_norm": 0.9233377575874329,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 6120
+    },
+    {
+      "epoch": 4.757469926270858,
+      "grad_norm": 0.9675140976905823,
+      "learning_rate": 0.0002,
+      "loss": 1.2642,
+      "step": 6130
+    },
+    {
+      "epoch": 4.765230888630191,
+      "grad_norm": 0.9639796018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6140
+    },
+    {
+      "epoch": 4.772991850989523,
+      "grad_norm": 0.925199568271637,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 6150
+    },
+    {
+      "epoch": 4.7807528133488555,
+      "grad_norm": 1.050901174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 6160
+    },
+    {
+      "epoch": 4.7885137757081875,
+      "grad_norm": 0.8920623660087585,
+      "learning_rate": 0.0002,
+      "loss": 1.301,
+      "step": 6170
+    },
+    {
+      "epoch": 4.79627473806752,
+      "grad_norm": 0.8964757919311523,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6180
+    },
+    {
+      "epoch": 4.804035700426853,
+      "grad_norm": 1.0839070081710815,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 6190
+    },
+    {
+      "epoch": 4.811796662786185,
+      "grad_norm": 0.8809942007064819,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 6200
+    },
+    {
+      "epoch": 4.819557625145518,
+      "grad_norm": 1.0216195583343506,
+      "learning_rate": 0.0002,
+      "loss": 1.321,
+      "step": 6210
+    },
+    {
+      "epoch": 4.827318587504851,
+      "grad_norm": 0.892005980014801,
+      "learning_rate": 0.0002,
+      "loss": 1.3033,
+      "step": 6220
+    },
+    {
+      "epoch": 4.835079549864183,
+      "grad_norm": 0.9957166910171509,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6230
+    },
+    {
+      "epoch": 4.842840512223516,
+      "grad_norm": 0.9720533490180969,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 6240
+    },
+    {
+      "epoch": 4.850601474582849,
+      "grad_norm": 0.9336182475090027,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 6250
+    },
+    {
+      "epoch": 4.858362436942181,
+      "grad_norm": 1.2611457109451294,
+      "learning_rate": 0.0002,
+      "loss": 1.3136,
+      "step": 6260
+    },
+    {
+      "epoch": 4.866123399301514,
+      "grad_norm": 0.8927203416824341,
+      "learning_rate": 0.0002,
+      "loss": 1.2234,
+      "step": 6270
+    },
+    {
+      "epoch": 4.873884361660846,
+      "grad_norm": 0.9706710577011108,
+      "learning_rate": 0.0002,
+      "loss": 1.3463,
+      "step": 6280
+    },
+    {
+      "epoch": 4.8816453240201785,
+      "grad_norm": 1.1461690664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 6290
+    },
+    {
+      "epoch": 4.889406286379511,
+      "grad_norm": 0.9930381178855896,
+      "learning_rate": 0.0002,
+      "loss": 1.2566,
+      "step": 6300
+    },
+    {
+      "epoch": 4.897167248738843,
+      "grad_norm": 0.91451096534729,
+      "learning_rate": 0.0002,
+      "loss": 1.2568,
+      "step": 6310
+    },
+    {
+      "epoch": 4.904928211098176,
+      "grad_norm": 1.0319571495056152,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 6320
+    },
+    {
+      "epoch": 4.912689173457509,
+      "grad_norm": 0.990140438079834,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6330
+    },
+    {
+      "epoch": 4.920450135816841,
+      "grad_norm": 1.2466117143630981,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 6340
+    },
+    {
+      "epoch": 4.928211098176174,
+      "grad_norm": 1.0316979885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 6350
+    },
+    {
+      "epoch": 4.935972060535507,
+      "grad_norm": 1.0643759965896606,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 6360
+    },
+    {
+      "epoch": 4.943733022894839,
+      "grad_norm": 0.9703279733657837,
+      "learning_rate": 0.0002,
+      "loss": 1.2559,
+      "step": 6370
+    },
+    {
+      "epoch": 4.951493985254172,
+      "grad_norm": 0.9767927527427673,
+      "learning_rate": 0.0002,
+      "loss": 1.2155,
+      "step": 6380
+    },
+    {
+      "epoch": 4.959254947613504,
+      "grad_norm": 0.960854172706604,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 6390
+    },
+    {
+      "epoch": 4.967015909972837,
+      "grad_norm": 0.9922910332679749,
+      "learning_rate": 0.0002,
+      "loss": 1.3314,
+      "step": 6400
+    },
+    {
+      "epoch": 4.9747768723321695,
+      "grad_norm": 0.956470787525177,
+      "learning_rate": 0.0002,
+      "loss": 1.3018,
+      "step": 6410
+    },
+    {
+      "epoch": 4.9825378346915015,
+      "grad_norm": 0.9637242555618286,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 6420
+    },
+    {
+      "epoch": 4.990298797050834,
+      "grad_norm": 1.0855202674865723,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 6430
+    },
+    {
+      "epoch": 4.998059759410167,
+      "grad_norm": 0.9655316472053528,
+      "learning_rate": 0.0002,
+      "loss": 1.3015,
+      "step": 6440
+    },
+    {
+      "epoch": 4.9996119518820334,
+      "eval_loss": 2.0410802364349365,
+      "eval_runtime": 113.04,
+      "eval_samples_per_second": 4.485,
+      "eval_steps_per_second": 0.566,
+      "step": 6442
+    },
+    {
+      "epoch": 5.005820721769499,
+      "grad_norm": 1.1676199436187744,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 6450
+    },
+    {
+      "epoch": 5.013581684128832,
+      "grad_norm": 1.4317965507507324,
+      "learning_rate": 0.0002,
+      "loss": 1.041,
+      "step": 6460
+    },
+    {
+      "epoch": 5.021342646488165,
+      "grad_norm": 1.460443377494812,
+      "learning_rate": 0.0002,
+      "loss": 0.9546,
+      "step": 6470
+    },
+    {
+      "epoch": 5.029103608847497,
+      "grad_norm": 1.2299214601516724,
+      "learning_rate": 0.0002,
+      "loss": 1.0014,
+      "step": 6480
+    },
+    {
+      "epoch": 5.03686457120683,
+      "grad_norm": 1.3125724792480469,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 6490
+    },
+    {
+      "epoch": 5.044625533566162,
+      "grad_norm": 1.1252319812774658,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 6500
+    },
+    {
+      "epoch": 5.052386495925495,
+      "grad_norm": 0.9970866441726685,
+      "learning_rate": 0.0002,
+      "loss": 0.976,
+      "step": 6510
+    },
+    {
+      "epoch": 5.060147458284828,
+      "grad_norm": 1.229069709777832,
+      "learning_rate": 0.0002,
+      "loss": 0.9731,
+      "step": 6520
+    },
+    {
+      "epoch": 5.06790842064416,
+      "grad_norm": 1.2430938482284546,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 6530
+    },
+    {
+      "epoch": 5.0756693830034925,
+      "grad_norm": 1.0522737503051758,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 6540
+    },
+    {
+      "epoch": 5.083430345362825,
+      "grad_norm": 1.108890175819397,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 6550
+    },
+    {
+      "epoch": 5.091191307722157,
+      "grad_norm": 1.156912922859192,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 6560
+    },
+    {
+      "epoch": 5.09895227008149,
+      "grad_norm": 1.405895709991455,
+      "learning_rate": 0.0002,
+      "loss": 1.0415,
+      "step": 6570
+    },
+    {
+      "epoch": 5.106713232440823,
+      "grad_norm": 1.2005155086517334,
+      "learning_rate": 0.0002,
+      "loss": 0.9811,
+      "step": 6580
+    },
+    {
+      "epoch": 5.114474194800155,
+      "grad_norm": 1.181443452835083,
+      "learning_rate": 0.0002,
+      "loss": 0.9862,
+      "step": 6590
+    },
+    {
+      "epoch": 5.122235157159488,
+      "grad_norm": 2.3444771766662598,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 6600
+    },
+    {
+      "epoch": 5.12999611951882,
+      "grad_norm": 1.216988444328308,
+      "learning_rate": 0.0002,
+      "loss": 1.0455,
+      "step": 6610
+    },
+    {
+      "epoch": 5.137757081878153,
+      "grad_norm": 1.369553565979004,
+      "learning_rate": 0.0002,
+      "loss": 1.0549,
+      "step": 6620
+    },
+    {
+      "epoch": 5.145518044237486,
+      "grad_norm": 1.177964687347412,
+      "learning_rate": 0.0002,
+      "loss": 1.0056,
+      "step": 6630
+    },
+    {
+      "epoch": 5.153279006596818,
+      "grad_norm": 1.1397041082382202,
+      "learning_rate": 0.0002,
+      "loss": 1.1025,
+      "step": 6640
+    },
+    {
+      "epoch": 5.161039968956151,
+      "grad_norm": 1.3976861238479614,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 6650
+    },
+    {
+      "epoch": 5.1688009313154835,
+      "grad_norm": 1.4824495315551758,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 6660
+    },
+    {
+      "epoch": 5.1765618936748155,
+      "grad_norm": 1.2653018236160278,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 6670
+    },
+    {
+      "epoch": 5.184322856034148,
+      "grad_norm": 1.3106069564819336,
+      "learning_rate": 0.0002,
+      "loss": 0.9971,
+      "step": 6680
+    },
+    {
+      "epoch": 5.192083818393481,
+      "grad_norm": 1.3140279054641724,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 6690
+    },
+    {
+      "epoch": 5.199844780752813,
+      "grad_norm": 1.3900256156921387,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 6700
+    },
+    {
+      "epoch": 5.207605743112146,
+      "grad_norm": 1.3191124200820923,
+      "learning_rate": 0.0002,
+      "loss": 1.0285,
+      "step": 6710
+    },
+    {
+      "epoch": 5.215366705471478,
+      "grad_norm": 1.176107406616211,
+      "learning_rate": 0.0002,
+      "loss": 0.9921,
+      "step": 6720
+    },
+    {
+      "epoch": 5.223127667830811,
+      "grad_norm": 1.2364883422851562,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 6730
+    },
+    {
+      "epoch": 5.230888630190144,
+      "grad_norm": 1.343022108078003,
+      "learning_rate": 0.0002,
+      "loss": 0.9599,
+      "step": 6740
+    },
+    {
+      "epoch": 5.238649592549476,
+      "grad_norm": 1.2826898097991943,
+      "learning_rate": 0.0002,
+      "loss": 1.0342,
+      "step": 6750
+    },
+    {
+      "epoch": 5.246410554908809,
+      "grad_norm": 1.500257134437561,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 6760
+    },
+    {
+      "epoch": 5.254171517268142,
+      "grad_norm": 1.2605743408203125,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 6770
+    },
+    {
+      "epoch": 5.261932479627474,
+      "grad_norm": 1.2355525493621826,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6780
+    },
+    {
+      "epoch": 5.2696934419868064,
+      "grad_norm": 1.2845789194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.0436,
+      "step": 6790
+    },
+    {
+      "epoch": 5.277454404346139,
+      "grad_norm": 1.3696625232696533,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 6800
+    },
+    {
+      "epoch": 5.285215366705471,
+      "grad_norm": 1.4051260948181152,
+      "learning_rate": 0.0002,
+      "loss": 1.0991,
+      "step": 6810
+    },
+    {
+      "epoch": 5.292976329064804,
+      "grad_norm": 1.266725778579712,
+      "learning_rate": 0.0002,
+      "loss": 1.0987,
+      "step": 6820
+    },
+    {
+      "epoch": 5.300737291424136,
+      "grad_norm": 1.3475236892700195,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 6830
+    },
+    {
+      "epoch": 5.308498253783469,
+      "grad_norm": 1.54409921169281,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 6840
+    },
+    {
+      "epoch": 5.316259216142802,
+      "grad_norm": 1.2391985654830933,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 6850
+    },
+    {
+      "epoch": 5.324020178502134,
+      "grad_norm": 1.2435699701309204,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 6860
+    },
+    {
+      "epoch": 5.331781140861467,
+      "grad_norm": 1.8803037405014038,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6870
+    },
+    {
+      "epoch": 5.3395421032208,
+      "grad_norm": 1.4195542335510254,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 6880
+    },
+    {
+      "epoch": 5.347303065580132,
+      "grad_norm": 1.1853394508361816,
+      "learning_rate": 0.0002,
+      "loss": 1.0273,
+      "step": 6890
+    },
+    {
+      "epoch": 5.355064027939465,
+      "grad_norm": 1.4016530513763428,
+      "learning_rate": 0.0002,
+      "loss": 1.0668,
+      "step": 6900
+    },
+    {
+      "epoch": 5.3628249902987974,
+      "grad_norm": 1.294339895248413,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 6910
+    },
+    {
+      "epoch": 5.370585952658129,
+      "grad_norm": 1.2952708005905151,
+      "learning_rate": 0.0002,
+      "loss": 1.0724,
+      "step": 6920
+    },
+    {
+      "epoch": 5.378346915017462,
+      "grad_norm": 1.1361510753631592,
+      "learning_rate": 0.0002,
+      "loss": 1.0098,
+      "step": 6930
+    },
+    {
+      "epoch": 5.386107877376794,
+      "grad_norm": 1.125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 6940
+    },
+    {
+      "epoch": 5.393868839736127,
+      "grad_norm": 1.1453300714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 6950
+    },
+    {
+      "epoch": 5.40162980209546,
+      "grad_norm": 1.4542768001556396,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6960
+    },
+    {
+      "epoch": 5.409390764454792,
+      "grad_norm": 1.2360988855361938,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6970
+    },
+    {
+      "epoch": 5.417151726814125,
+      "grad_norm": 1.2182754278182983,
+      "learning_rate": 0.0002,
+      "loss": 1.0631,
+      "step": 6980
+    },
+    {
+      "epoch": 5.424912689173458,
+      "grad_norm": 1.2018693685531616,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 6990
+    },
+    {
+      "epoch": 5.43267365153279,
+      "grad_norm": 1.346124291419983,
+      "learning_rate": 0.0002,
+      "loss": 1.108,
+      "step": 7000
+    },
+    {
+      "epoch": 5.440434613892123,
+      "grad_norm": 1.2534189224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7010
+    },
+    {
+      "epoch": 5.448195576251456,
+      "grad_norm": 1.2033339738845825,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 7020
+    },
+    {
+      "epoch": 5.4559565386107876,
+      "grad_norm": 1.2788134813308716,
+      "learning_rate": 0.0002,
+      "loss": 1.0714,
+      "step": 7030
+    },
+    {
+      "epoch": 5.46371750097012,
+      "grad_norm": 1.2751542329788208,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 7040
+    },
+    {
+      "epoch": 5.471478463329452,
+      "grad_norm": 1.3237019777297974,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7050
+    },
+    {
+      "epoch": 5.479239425688785,
+      "grad_norm": 1.4932852983474731,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 7060
+    },
+    {
+      "epoch": 5.487000388048118,
+      "grad_norm": 1.4003876447677612,
+      "learning_rate": 0.0002,
+      "loss": 1.0197,
+      "step": 7070
+    },
+    {
+      "epoch": 5.49476135040745,
+      "grad_norm": 1.404799461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7080
+    },
+    {
+      "epoch": 5.502522312766783,
+      "grad_norm": 1.4486982822418213,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 7090
+    },
+    {
+      "epoch": 5.510283275126116,
+      "grad_norm": 1.1713480949401855,
+      "learning_rate": 0.0002,
+      "loss": 1.0645,
+      "step": 7100
+    },
+    {
+      "epoch": 5.518044237485448,
+      "grad_norm": 1.4062601327896118,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 7110
+    },
+    {
+      "epoch": 5.525805199844781,
+      "grad_norm": 1.211629867553711,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 7120
+    },
+    {
+      "epoch": 5.533566162204114,
+      "grad_norm": 1.2523176670074463,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 7130
+    },
+    {
+      "epoch": 5.541327124563446,
+      "grad_norm": 1.4467198848724365,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 7140
+    },
+    {
+      "epoch": 5.5490880869227786,
+      "grad_norm": 1.5961614847183228,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 7150
+    },
+    {
+      "epoch": 5.5568490492821105,
+      "grad_norm": 1.320656418800354,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 7160
+    },
+    {
+      "epoch": 5.564610011641443,
+      "grad_norm": 1.2423332929611206,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 7170
+    },
+    {
+      "epoch": 5.572370974000776,
+      "grad_norm": 1.2919669151306152,
+      "learning_rate": 0.0002,
+      "loss": 1.0046,
+      "step": 7180
+    },
+    {
+      "epoch": 5.580131936360108,
+      "grad_norm": 1.1678385734558105,
+      "learning_rate": 0.0002,
+      "loss": 1.046,
+      "step": 7190
+    },
+    {
+      "epoch": 5.587892898719441,
+      "grad_norm": 1.4250764846801758,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 7200
+    },
+    {
+      "epoch": 5.595653861078774,
+      "grad_norm": 1.5308716297149658,
+      "learning_rate": 0.0002,
+      "loss": 1.1254,
+      "step": 7210
+    },
+    {
+      "epoch": 5.603414823438106,
+      "grad_norm": 1.2678815126419067,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7220
+    },
+    {
+      "epoch": 5.611175785797439,
+      "grad_norm": 1.127856969833374,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 7230
+    },
+    {
+      "epoch": 5.618936748156772,
+      "grad_norm": 1.3832560777664185,
+      "learning_rate": 0.0002,
+      "loss": 1.0647,
+      "step": 7240
+    },
+    {
+      "epoch": 5.626697710516104,
+      "grad_norm": 1.3226919174194336,
+      "learning_rate": 0.0002,
+      "loss": 1.0658,
+      "step": 7250
+    },
+    {
+      "epoch": 5.634458672875437,
+      "grad_norm": 1.3418006896972656,
+      "learning_rate": 0.0002,
+      "loss": 1.1175,
+      "step": 7260
+    },
+    {
+      "epoch": 5.642219635234769,
+      "grad_norm": 1.2625300884246826,
+      "learning_rate": 0.0002,
+      "loss": 1.0956,
+      "step": 7270
+    },
+    {
+      "epoch": 5.6499805975941015,
+      "grad_norm": 1.1579464673995972,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 7280
+    },
+    {
+      "epoch": 5.657741559953434,
+      "grad_norm": 1.4998650550842285,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 7290
+    },
+    {
+      "epoch": 5.665502522312766,
+      "grad_norm": 1.2670758962631226,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7300
+    },
+    {
+      "epoch": 5.673263484672099,
+      "grad_norm": 1.2959760427474976,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 7310
+    },
+    {
+      "epoch": 5.681024447031432,
+      "grad_norm": 1.2460671663284302,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 7320
+    },
+    {
+      "epoch": 5.688785409390764,
+      "grad_norm": 1.1313989162445068,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 7330
+    },
+    {
+      "epoch": 5.696546371750097,
+      "grad_norm": 1.282527208328247,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 7340
+    },
+    {
+      "epoch": 5.70430733410943,
+      "grad_norm": 1.3380206823349,
+      "learning_rate": 0.0002,
+      "loss": 1.1315,
+      "step": 7350
+    },
+    {
+      "epoch": 5.712068296468762,
+      "grad_norm": 1.1648279428482056,
+      "learning_rate": 0.0002,
+      "loss": 1.0949,
+      "step": 7360
+    },
+    {
+      "epoch": 5.719829258828095,
+      "grad_norm": 1.3059816360473633,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 7370
+    },
+    {
+      "epoch": 5.727590221187427,
+      "grad_norm": 1.1905046701431274,
+      "learning_rate": 0.0002,
+      "loss": 1.1496,
+      "step": 7380
+    },
+    {
+      "epoch": 5.73535118354676,
+      "grad_norm": 1.4089630842208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1356,
+      "step": 7390
+    },
+    {
+      "epoch": 5.7431121459060925,
+      "grad_norm": 1.256721019744873,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 7400
+    },
+    {
+      "epoch": 5.7508731082654245,
+      "grad_norm": 1.1915162801742554,
+      "learning_rate": 0.0002,
+      "loss": 1.0682,
+      "step": 7410
+    },
+    {
+      "epoch": 5.758634070624757,
+      "grad_norm": 1.1935480833053589,
+      "learning_rate": 0.0002,
+      "loss": 1.1257,
+      "step": 7420
+    },
+    {
+      "epoch": 5.76639503298409,
+      "grad_norm": 1.1761008501052856,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 7430
+    },
+    {
+      "epoch": 5.774155995343422,
+      "grad_norm": 1.2540549039840698,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 7440
+    },
+    {
+      "epoch": 5.781916957702755,
+      "grad_norm": 1.5295120477676392,
+      "learning_rate": 0.0002,
+      "loss": 1.1527,
+      "step": 7450
+    },
+    {
+      "epoch": 5.789677920062088,
+      "grad_norm": 1.1081160306930542,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 7460
+    },
+    {
+      "epoch": 5.79743888242142,
+      "grad_norm": 1.4381253719329834,
+      "learning_rate": 0.0002,
+      "loss": 1.1304,
+      "step": 7470
+    },
+    {
+      "epoch": 5.805199844780753,
+      "grad_norm": 1.3079341650009155,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 7480
+    },
+    {
+      "epoch": 5.812960807140085,
+      "grad_norm": 1.1372792720794678,
+      "learning_rate": 0.0002,
+      "loss": 1.0544,
+      "step": 7490
+    },
+    {
+      "epoch": 5.820721769499418,
+      "grad_norm": 1.3221744298934937,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 7500
+    },
+    {
+      "epoch": 5.828482731858751,
+      "grad_norm": 1.3436939716339111,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 7510
+    },
+    {
+      "epoch": 5.8362436942180835,
+      "grad_norm": 1.3916879892349243,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 7520
+    },
+    {
+      "epoch": 5.8440046565774155,
+      "grad_norm": 1.2463704347610474,
+      "learning_rate": 0.0002,
+      "loss": 1.0816,
+      "step": 7530
+    },
+    {
+      "epoch": 5.851765618936748,
+      "grad_norm": 1.097051739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.0745,
+      "step": 7540
+    },
+    {
+      "epoch": 5.85952658129608,
+      "grad_norm": 1.1554739475250244,
+      "learning_rate": 0.0002,
+      "loss": 1.1454,
+      "step": 7550
+    },
+    {
+      "epoch": 5.867287543655413,
+      "grad_norm": 1.2384694814682007,
+      "learning_rate": 0.0002,
+      "loss": 1.0953,
+      "step": 7560
+    },
+    {
+      "epoch": 5.875048506014746,
+      "grad_norm": 1.142815351486206,
+      "learning_rate": 0.0002,
+      "loss": 1.1734,
+      "step": 7570
+    },
+    {
+      "epoch": 5.882809468374078,
+      "grad_norm": 1.3637062311172485,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 7580
+    },
+    {
+      "epoch": 5.890570430733411,
+      "grad_norm": 1.2449073791503906,
+      "learning_rate": 0.0002,
+      "loss": 1.0781,
+      "step": 7590
+    },
+    {
+      "epoch": 5.898331393092743,
+      "grad_norm": 1.358058214187622,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 7600
+    },
+    {
+      "epoch": 5.906092355452076,
+      "grad_norm": 1.264655351638794,
+      "learning_rate": 0.0002,
+      "loss": 1.0779,
+      "step": 7610
+    },
+    {
+      "epoch": 5.913853317811409,
+      "grad_norm": 1.3186019659042358,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 7620
+    },
+    {
+      "epoch": 5.921614280170742,
+      "grad_norm": 1.4111460447311401,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 7630
+    },
+    {
+      "epoch": 5.929375242530074,
+      "grad_norm": 1.1078972816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.1765,
+      "step": 7640
+    },
+    {
+      "epoch": 5.9371362048894065,
+      "grad_norm": 1.2742213010787964,
+      "learning_rate": 0.0002,
+      "loss": 1.1305,
+      "step": 7650
+    },
+    {
+      "epoch": 5.9448971672487385,
+      "grad_norm": 1.3412781953811646,
+      "learning_rate": 0.0002,
+      "loss": 1.144,
+      "step": 7660
+    },
+    {
+      "epoch": 5.952658129608071,
+      "grad_norm": 1.123005986213684,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 7670
+    },
+    {
+      "epoch": 5.960419091967404,
+      "grad_norm": 1.2203444242477417,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 7680
+    },
+    {
+      "epoch": 5.968180054326736,
+      "grad_norm": 1.341011643409729,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 7690
+    },
+    {
+      "epoch": 5.975941016686069,
+      "grad_norm": 1.2689454555511475,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 7700
+    },
+    {
+      "epoch": 5.983701979045401,
+      "grad_norm": 1.1518112421035767,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 7710
+    },
+    {
+      "epoch": 5.991462941404734,
+      "grad_norm": 1.3698320388793945,
+      "learning_rate": 0.0002,
+      "loss": 1.1868,
+      "step": 7720
+    },
+    {
+      "epoch": 5.999223903764067,
+      "grad_norm": 1.2812788486480713,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 7730
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.252762794494629,
+      "eval_runtime": 114.8471,
+      "eval_samples_per_second": 4.415,
+      "eval_steps_per_second": 0.557,
+      "step": 7731
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 3.9709311745130496e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-7731/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+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/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e98db163734cc03f7a8f8b3f720d3a2befdf7453
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "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": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..332e89924059813e5123e5202e68195f613696ea
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc1031f2b40f599415c82b304b1942ecafef76a9222d3be72dcb17285197c566
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e68c07c24b855908a4caf1ec5ed18fcd10eff2e9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0e4e2adac8c101bbb66ef23229310b1ce252a5375c7c85126c484fc3582cb29
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fe38c419ee36480fe8ade37f82697fddcec3abda
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70c3bb1cda34731925ef8ba2681db4519973fb310959491b41b809191c7954c5
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..903fde5a3bc95e50951322f409e90b32a9316276
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2486a6be3f3aae5c8d316423c16cad160bed976bd968bd54f8c859cc457b20c
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..af662405c44612d20d5861526d5f22c5e1feb8d9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/trainer_state.json
@@ -0,0 +1,6396 @@
+{
+  "best_metric": 1.8068748712539673,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577",
+  "epoch": 6.9996119518820334,
+  "eval_steps": 10,
+  "global_step": 9019,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007760962359332557,
+      "grad_norm": 1.0751162767410278,
+      "learning_rate": 0.0002,
+      "loss": 3.0855,
+      "step": 10
+    },
+    {
+      "epoch": 0.015521924718665115,
+      "grad_norm": 0.4697345793247223,
+      "learning_rate": 0.0002,
+      "loss": 2.4744,
+      "step": 20
+    },
+    {
+      "epoch": 0.023282887077997673,
+      "grad_norm": 0.5370839238166809,
+      "learning_rate": 0.0002,
+      "loss": 2.193,
+      "step": 30
+    },
+    {
+      "epoch": 0.03104384943733023,
+      "grad_norm": 0.46794816851615906,
+      "learning_rate": 0.0002,
+      "loss": 2.0599,
+      "step": 40
+    },
+    {
+      "epoch": 0.038804811796662786,
+      "grad_norm": 0.44624820351600647,
+      "learning_rate": 0.0002,
+      "loss": 1.9354,
+      "step": 50
+    },
+    {
+      "epoch": 0.046565774155995346,
+      "grad_norm": 0.3953201472759247,
+      "learning_rate": 0.0002,
+      "loss": 1.9319,
+      "step": 60
+    },
+    {
+      "epoch": 0.0543267365153279,
+      "grad_norm": 0.3935912549495697,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 70
+    },
+    {
+      "epoch": 0.06208769887466046,
+      "grad_norm": 0.4520699381828308,
+      "learning_rate": 0.0002,
+      "loss": 1.8795,
+      "step": 80
+    },
+    {
+      "epoch": 0.06984866123399301,
+      "grad_norm": 0.3801847994327545,
+      "learning_rate": 0.0002,
+      "loss": 1.8354,
+      "step": 90
+    },
+    {
+      "epoch": 0.07760962359332557,
+      "grad_norm": 0.4020165205001831,
+      "learning_rate": 0.0002,
+      "loss": 1.9053,
+      "step": 100
+    },
+    {
+      "epoch": 0.08537058595265813,
+      "grad_norm": 0.3860672116279602,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 110
+    },
+    {
+      "epoch": 0.09313154831199069,
+      "grad_norm": 0.3681113123893738,
+      "learning_rate": 0.0002,
+      "loss": 1.8731,
+      "step": 120
+    },
+    {
+      "epoch": 0.10089251067132324,
+      "grad_norm": 0.3594866991043091,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 130
+    },
+    {
+      "epoch": 0.1086534730306558,
+      "grad_norm": 0.3879193663597107,
+      "learning_rate": 0.0002,
+      "loss": 1.8266,
+      "step": 140
+    },
+    {
+      "epoch": 0.11641443538998836,
+      "grad_norm": 0.3270505666732788,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 150
+    },
+    {
+      "epoch": 0.12417539774932092,
+      "grad_norm": 0.36824458837509155,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 160
+    },
+    {
+      "epoch": 0.13193636010865348,
+      "grad_norm": 0.383882075548172,
+      "learning_rate": 0.0002,
+      "loss": 1.8305,
+      "step": 170
+    },
+    {
+      "epoch": 0.13969732246798602,
+      "grad_norm": 0.3368665874004364,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 180
+    },
+    {
+      "epoch": 0.1474582848273186,
+      "grad_norm": 0.35961097478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.7882,
+      "step": 190
+    },
+    {
+      "epoch": 0.15521924718665114,
+      "grad_norm": 0.3415963351726532,
+      "learning_rate": 0.0002,
+      "loss": 1.8467,
+      "step": 200
+    },
+    {
+      "epoch": 0.1629802095459837,
+      "grad_norm": 0.4100632071495056,
+      "learning_rate": 0.0002,
+      "loss": 1.8543,
+      "step": 210
+    },
+    {
+      "epoch": 0.17074117190531626,
+      "grad_norm": 0.3516307473182678,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 220
+    },
+    {
+      "epoch": 0.1785021342646488,
+      "grad_norm": 0.37919050455093384,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 230
+    },
+    {
+      "epoch": 0.18626309662398138,
+      "grad_norm": 0.33270683884620667,
+      "learning_rate": 0.0002,
+      "loss": 1.7937,
+      "step": 240
+    },
+    {
+      "epoch": 0.19402405898331393,
+      "grad_norm": 0.3348783254623413,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 250
+    },
+    {
+      "epoch": 0.20178502134264648,
+      "grad_norm": 0.3888475298881531,
+      "learning_rate": 0.0002,
+      "loss": 1.7774,
+      "step": 260
+    },
+    {
+      "epoch": 0.20954598370197905,
+      "grad_norm": 0.3554602861404419,
+      "learning_rate": 0.0002,
+      "loss": 1.8381,
+      "step": 270
+    },
+    {
+      "epoch": 0.2173069460613116,
+      "grad_norm": 0.33277708292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8359,
+      "step": 280
+    },
+    {
+      "epoch": 0.22506790842064417,
+      "grad_norm": 0.3281584680080414,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 290
+    },
+    {
+      "epoch": 0.23282887077997672,
+      "grad_norm": 0.3185969591140747,
+      "learning_rate": 0.0002,
+      "loss": 1.8181,
+      "step": 300
+    },
+    {
+      "epoch": 0.24058983313930926,
+      "grad_norm": 0.35335442423820496,
+      "learning_rate": 0.0002,
+      "loss": 1.8595,
+      "step": 310
+    },
+    {
+      "epoch": 0.24835079549864184,
+      "grad_norm": 0.3119595944881439,
+      "learning_rate": 0.0002,
+      "loss": 1.87,
+      "step": 320
+    },
+    {
+      "epoch": 0.2561117578579744,
+      "grad_norm": 0.36424458026885986,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 330
+    },
+    {
+      "epoch": 0.26387272021730696,
+      "grad_norm": 0.3618951141834259,
+      "learning_rate": 0.0002,
+      "loss": 1.8003,
+      "step": 340
+    },
+    {
+      "epoch": 0.2716336825766395,
+      "grad_norm": 0.312757670879364,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 350
+    },
+    {
+      "epoch": 0.27939464493597205,
+      "grad_norm": 0.326016366481781,
+      "learning_rate": 0.0002,
+      "loss": 1.9031,
+      "step": 360
+    },
+    {
+      "epoch": 0.2871556072953046,
+      "grad_norm": 0.34093883633613586,
+      "learning_rate": 0.0002,
+      "loss": 1.8214,
+      "step": 370
+    },
+    {
+      "epoch": 0.2949165696546372,
+      "grad_norm": 0.32325029373168945,
+      "learning_rate": 0.0002,
+      "loss": 1.7733,
+      "step": 380
+    },
+    {
+      "epoch": 0.30267753201396974,
+      "grad_norm": 0.34105437994003296,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 390
+    },
+    {
+      "epoch": 0.3104384943733023,
+      "grad_norm": 0.32565295696258545,
+      "learning_rate": 0.0002,
+      "loss": 1.7926,
+      "step": 400
+    },
+    {
+      "epoch": 0.31819945673263483,
+      "grad_norm": 0.32742050290107727,
+      "learning_rate": 0.0002,
+      "loss": 1.8031,
+      "step": 410
+    },
+    {
+      "epoch": 0.3259604190919674,
+      "grad_norm": 0.30233046412467957,
+      "learning_rate": 0.0002,
+      "loss": 1.907,
+      "step": 420
+    },
+    {
+      "epoch": 0.3337213814513,
+      "grad_norm": 0.32419222593307495,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 430
+    },
+    {
+      "epoch": 0.3414823438106325,
+      "grad_norm": 0.3653007745742798,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 440
+    },
+    {
+      "epoch": 0.3492433061699651,
+      "grad_norm": 0.31617099046707153,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 450
+    },
+    {
+      "epoch": 0.3570042685292976,
+      "grad_norm": 0.3305962085723877,
+      "learning_rate": 0.0002,
+      "loss": 1.7677,
+      "step": 460
+    },
+    {
+      "epoch": 0.36476523088863017,
+      "grad_norm": 0.3178933262825012,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 470
+    },
+    {
+      "epoch": 0.37252619324796277,
+      "grad_norm": 0.37163782119750977,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 480
+    },
+    {
+      "epoch": 0.3802871556072953,
+      "grad_norm": 0.469844788312912,
+      "learning_rate": 0.0002,
+      "loss": 1.8804,
+      "step": 490
+    },
+    {
+      "epoch": 0.38804811796662786,
+      "grad_norm": 0.3409338593482971,
+      "learning_rate": 0.0002,
+      "loss": 1.8343,
+      "step": 500
+    },
+    {
+      "epoch": 0.3958090803259604,
+      "grad_norm": 0.31943467259407043,
+      "learning_rate": 0.0002,
+      "loss": 1.8433,
+      "step": 510
+    },
+    {
+      "epoch": 0.40357004268529295,
+      "grad_norm": 0.32293614745140076,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 520
+    },
+    {
+      "epoch": 0.41133100504462555,
+      "grad_norm": 0.2994382977485657,
+      "learning_rate": 0.0002,
+      "loss": 1.8584,
+      "step": 530
+    },
+    {
+      "epoch": 0.4190919674039581,
+      "grad_norm": 0.3273141384124756,
+      "learning_rate": 0.0002,
+      "loss": 1.8153,
+      "step": 540
+    },
+    {
+      "epoch": 0.42685292976329064,
+      "grad_norm": 0.3020550012588501,
+      "learning_rate": 0.0002,
+      "loss": 1.8097,
+      "step": 550
+    },
+    {
+      "epoch": 0.4346138921226232,
+      "grad_norm": 0.30113112926483154,
+      "learning_rate": 0.0002,
+      "loss": 1.8551,
+      "step": 560
+    },
+    {
+      "epoch": 0.44237485448195574,
+      "grad_norm": 0.30274903774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 570
+    },
+    {
+      "epoch": 0.45013581684128834,
+      "grad_norm": 0.3231128454208374,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 580
+    },
+    {
+      "epoch": 0.4578967792006209,
+      "grad_norm": 0.3255121409893036,
+      "learning_rate": 0.0002,
+      "loss": 1.7848,
+      "step": 590
+    },
+    {
+      "epoch": 0.46565774155995343,
+      "grad_norm": 0.30147507786750793,
+      "learning_rate": 0.0002,
+      "loss": 1.8227,
+      "step": 600
+    },
+    {
+      "epoch": 0.473418703919286,
+      "grad_norm": 0.29781386256217957,
+      "learning_rate": 0.0002,
+      "loss": 1.7572,
+      "step": 610
+    },
+    {
+      "epoch": 0.4811796662786185,
+      "grad_norm": 0.30914685130119324,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 620
+    },
+    {
+      "epoch": 0.4889406286379511,
+      "grad_norm": 0.3110593855381012,
+      "learning_rate": 0.0002,
+      "loss": 1.805,
+      "step": 630
+    },
+    {
+      "epoch": 0.49670159099728367,
+      "grad_norm": 0.3298132121562958,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 640
+    },
+    {
+      "epoch": 0.5044625533566163,
+      "grad_norm": 0.322122186422348,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 650
+    },
+    {
+      "epoch": 0.5122235157159488,
+      "grad_norm": 0.3504371643066406,
+      "learning_rate": 0.0002,
+      "loss": 1.8001,
+      "step": 660
+    },
+    {
+      "epoch": 0.5199844780752814,
+      "grad_norm": 0.3102182149887085,
+      "learning_rate": 0.0002,
+      "loss": 1.8682,
+      "step": 670
+    },
+    {
+      "epoch": 0.5277454404346139,
+      "grad_norm": 0.6113658547401428,
+      "learning_rate": 0.0002,
+      "loss": 1.7494,
+      "step": 680
+    },
+    {
+      "epoch": 0.5355064027939465,
+      "grad_norm": 0.31841862201690674,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 690
+    },
+    {
+      "epoch": 0.543267365153279,
+      "grad_norm": 0.2830526530742645,
+      "learning_rate": 0.0002,
+      "loss": 1.7587,
+      "step": 700
+    },
+    {
+      "epoch": 0.5510283275126115,
+      "grad_norm": 0.3048769533634186,
+      "learning_rate": 0.0002,
+      "loss": 1.7887,
+      "step": 710
+    },
+    {
+      "epoch": 0.5587892898719441,
+      "grad_norm": 0.2719033658504486,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 720
+    },
+    {
+      "epoch": 0.5665502522312766,
+      "grad_norm": 0.3176722526550293,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 730
+    },
+    {
+      "epoch": 0.5743112145906092,
+      "grad_norm": 0.32491734623908997,
+      "learning_rate": 0.0002,
+      "loss": 1.7127,
+      "step": 740
+    },
+    {
+      "epoch": 0.5820721769499418,
+      "grad_norm": 0.32746851444244385,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 750
+    },
+    {
+      "epoch": 0.5898331393092744,
+      "grad_norm": 0.3055773973464966,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 760
+    },
+    {
+      "epoch": 0.5975941016686069,
+      "grad_norm": 0.30671584606170654,
+      "learning_rate": 0.0002,
+      "loss": 1.8597,
+      "step": 770
+    },
+    {
+      "epoch": 0.6053550640279395,
+      "grad_norm": 0.28770264983177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 780
+    },
+    {
+      "epoch": 0.613116026387272,
+      "grad_norm": 0.2814285457134247,
+      "learning_rate": 0.0002,
+      "loss": 1.7025,
+      "step": 790
+    },
+    {
+      "epoch": 0.6208769887466046,
+      "grad_norm": 0.31554412841796875,
+      "learning_rate": 0.0002,
+      "loss": 1.819,
+      "step": 800
+    },
+    {
+      "epoch": 0.6286379511059371,
+      "grad_norm": 0.2984226942062378,
+      "learning_rate": 0.0002,
+      "loss": 1.8335,
+      "step": 810
+    },
+    {
+      "epoch": 0.6363989134652697,
+      "grad_norm": 0.2859906554222107,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 820
+    },
+    {
+      "epoch": 0.6441598758246022,
+      "grad_norm": 0.2887928783893585,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 830
+    },
+    {
+      "epoch": 0.6519208381839348,
+      "grad_norm": 0.31287339329719543,
+      "learning_rate": 0.0002,
+      "loss": 1.8071,
+      "step": 840
+    },
+    {
+      "epoch": 0.6596818005432674,
+      "grad_norm": 0.32064181566238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 850
+    },
+    {
+      "epoch": 0.6674427629026,
+      "grad_norm": 0.290981650352478,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 860
+    },
+    {
+      "epoch": 0.6752037252619325,
+      "grad_norm": 0.33060121536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.8089,
+      "step": 870
+    },
+    {
+      "epoch": 0.682964687621265,
+      "grad_norm": 0.27032899856567383,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 880
+    },
+    {
+      "epoch": 0.6907256499805976,
+      "grad_norm": 0.29031234979629517,
+      "learning_rate": 0.0002,
+      "loss": 1.8423,
+      "step": 890
+    },
+    {
+      "epoch": 0.6984866123399301,
+      "grad_norm": 0.2845142185688019,
+      "learning_rate": 0.0002,
+      "loss": 1.835,
+      "step": 900
+    },
+    {
+      "epoch": 0.7062475746992627,
+      "grad_norm": 0.8638312816619873,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 910
+    },
+    {
+      "epoch": 0.7140085370585952,
+      "grad_norm": 0.3086668848991394,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 920
+    },
+    {
+      "epoch": 0.7217694994179278,
+      "grad_norm": 0.2724177837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 930
+    },
+    {
+      "epoch": 0.7295304617772603,
+      "grad_norm": 0.289559006690979,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 940
+    },
+    {
+      "epoch": 0.737291424136593,
+      "grad_norm": 0.3000658452510834,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 950
+    },
+    {
+      "epoch": 0.7450523864959255,
+      "grad_norm": 0.33544042706489563,
+      "learning_rate": 0.0002,
+      "loss": 1.7736,
+      "step": 960
+    },
+    {
+      "epoch": 0.7528133488552581,
+      "grad_norm": 0.28593236207962036,
+      "learning_rate": 0.0002,
+      "loss": 1.6979,
+      "step": 970
+    },
+    {
+      "epoch": 0.7605743112145906,
+      "grad_norm": 0.313634991645813,
+      "learning_rate": 0.0002,
+      "loss": 1.8583,
+      "step": 980
+    },
+    {
+      "epoch": 0.7683352735739232,
+      "grad_norm": 0.2949385941028595,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 990
+    },
+    {
+      "epoch": 0.7760962359332557,
+      "grad_norm": 0.2920108437538147,
+      "learning_rate": 0.0002,
+      "loss": 1.8689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7838571982925883,
+      "grad_norm": 0.3245100677013397,
+      "learning_rate": 0.0002,
+      "loss": 1.8401,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7916181606519208,
+      "grad_norm": 0.3007619380950928,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7993791230112534,
+      "grad_norm": 0.3630852997303009,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8071400853705859,
+      "grad_norm": 0.2856379747390747,
+      "learning_rate": 0.0002,
+      "loss": 1.7655,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8149010477299186,
+      "grad_norm": 0.32476478815078735,
+      "learning_rate": 0.0002,
+      "loss": 1.8371,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8226620100892511,
+      "grad_norm": 0.5162565112113953,
+      "learning_rate": 0.0002,
+      "loss": 1.8039,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8304229724485837,
+      "grad_norm": 0.316496342420578,
+      "learning_rate": 0.0002,
+      "loss": 1.8862,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8381839348079162,
+      "grad_norm": 0.31977516412734985,
+      "learning_rate": 0.0002,
+      "loss": 1.8023,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8459448971672487,
+      "grad_norm": 0.269509494304657,
+      "learning_rate": 0.0002,
+      "loss": 1.8547,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8537058595265813,
+      "grad_norm": 0.31621453166007996,
+      "learning_rate": 0.0002,
+      "loss": 1.7811,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8614668218859138,
+      "grad_norm": 0.2946535050868988,
+      "learning_rate": 0.0002,
+      "loss": 1.739,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8692277842452464,
+      "grad_norm": 0.3088909983634949,
+      "learning_rate": 0.0002,
+      "loss": 1.7511,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8769887466045789,
+      "grad_norm": 0.33033716678619385,
+      "learning_rate": 0.0002,
+      "loss": 1.8228,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8847497089639115,
+      "grad_norm": 0.2954833507537842,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8925106713232441,
+      "grad_norm": 0.2950248122215271,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9002716336825767,
+      "grad_norm": 0.296661913394928,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9080325960419092,
+      "grad_norm": 0.35451310873031616,
+      "learning_rate": 0.0002,
+      "loss": 1.7967,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9157935584012418,
+      "grad_norm": 0.32705947756767273,
+      "learning_rate": 0.0002,
+      "loss": 1.8202,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9235545207605743,
+      "grad_norm": 0.3333960771560669,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9313154831199069,
+      "grad_norm": 0.3042232096195221,
+      "learning_rate": 0.0002,
+      "loss": 1.7801,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9390764454792394,
+      "grad_norm": 0.281553715467453,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1210
+    },
+    {
+      "epoch": 0.946837407838572,
+      "grad_norm": 0.3096391558647156,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9545983701979045,
+      "grad_norm": 0.2866271734237671,
+      "learning_rate": 0.0002,
+      "loss": 1.7401,
+      "step": 1230
+    },
+    {
+      "epoch": 0.962359332557237,
+      "grad_norm": 0.28394097089767456,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9701202949165697,
+      "grad_norm": 0.3249266743659973,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9778812572759022,
+      "grad_norm": 0.2896869480609894,
+      "learning_rate": 0.0002,
+      "loss": 1.7563,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9856422196352348,
+      "grad_norm": 0.29224586486816406,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9934031819945673,
+      "grad_norm": 0.2820223569869995,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9996119518820333,
+      "eval_loss": 1.8081045150756836,
+      "eval_runtime": 102.3056,
+      "eval_samples_per_second": 4.956,
+      "eval_steps_per_second": 0.626,
+      "step": 1288
+    },
+    {
+      "epoch": 1.0011641443538999,
+      "grad_norm": 0.3282551169395447,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0089251067132325,
+      "grad_norm": 0.30217495560646057,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1300
+    },
+    {
+      "epoch": 1.016686069072565,
+      "grad_norm": 0.30801767110824585,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1310
+    },
+    {
+      "epoch": 1.0244470314318976,
+      "grad_norm": 0.31816792488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.7756,
+      "step": 1320
+    },
+    {
+      "epoch": 1.03220799379123,
+      "grad_norm": 0.27794334292411804,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 1330
+    },
+    {
+      "epoch": 1.0399689561505627,
+      "grad_norm": 0.3018926680088043,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0477299185098952,
+      "grad_norm": 0.3552975356578827,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0554908808692278,
+      "grad_norm": 0.32590144872665405,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0632518432285603,
+      "grad_norm": 0.3435460925102234,
+      "learning_rate": 0.0002,
+      "loss": 1.6479,
+      "step": 1370
+    },
+    {
+      "epoch": 1.071012805587893,
+      "grad_norm": 0.35037797689437866,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0787737679472253,
+      "grad_norm": 0.31398263573646545,
+      "learning_rate": 0.0002,
+      "loss": 1.7868,
+      "step": 1390
+    },
+    {
+      "epoch": 1.086534730306558,
+      "grad_norm": 0.3134010434150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6729,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0942956926658907,
+      "grad_norm": 0.4599704444408417,
+      "learning_rate": 0.0002,
+      "loss": 1.751,
+      "step": 1410
+    },
+    {
+      "epoch": 1.102056655025223,
+      "grad_norm": 0.35852891206741333,
+      "learning_rate": 0.0002,
+      "loss": 1.6871,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1098176173845558,
+      "grad_norm": 0.35628634691238403,
+      "learning_rate": 0.0002,
+      "loss": 1.7083,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1175785797438882,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.6166,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1253395421032208,
+      "grad_norm": 1.3712416887283325,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1331005044625533,
+      "grad_norm": 0.38406670093536377,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 1460
+    },
+    {
+      "epoch": 1.140861466821886,
+      "grad_norm": 0.3402116000652313,
+      "learning_rate": 0.0002,
+      "loss": 1.7104,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1486224291812184,
+      "grad_norm": 0.341189444065094,
+      "learning_rate": 0.0002,
+      "loss": 1.7074,
+      "step": 1480
+    },
+    {
+      "epoch": 1.156383391540551,
+      "grad_norm": 0.36629995703697205,
+      "learning_rate": 0.0002,
+      "loss": 1.6468,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1641443538998835,
+      "grad_norm": 0.3499569296836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6952,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1719053162592161,
+      "grad_norm": 0.3663063943386078,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1796662786185488,
+      "grad_norm": 0.34851500391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1874272409778812,
+      "grad_norm": 0.35071656107902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6092,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1951882033372139,
+      "grad_norm": 0.42783796787261963,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2029491656965463,
+      "grad_norm": 0.31830692291259766,
+      "learning_rate": 0.0002,
+      "loss": 1.7499,
+      "step": 1550
+    },
+    {
+      "epoch": 1.210710128055879,
+      "grad_norm": 0.3597424626350403,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2184710904152114,
+      "grad_norm": 0.35233765840530396,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 1570
+    },
+    {
+      "epoch": 1.226232052774544,
+      "grad_norm": 0.35942912101745605,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2339930151338767,
+      "grad_norm": 0.36159393191337585,
+      "learning_rate": 0.0002,
+      "loss": 1.6598,
+      "step": 1590
+    },
+    {
+      "epoch": 1.2417539774932091,
+      "grad_norm": 0.3328469693660736,
+      "learning_rate": 0.0002,
+      "loss": 1.6697,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2495149398525418,
+      "grad_norm": 0.3089476525783539,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2572759022118742,
+      "grad_norm": 0.30947765707969666,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 1620
+    },
+    {
+      "epoch": 1.265036864571207,
+      "grad_norm": 0.32154011726379395,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2727978269305393,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.0002,
+      "loss": 1.6621,
+      "step": 1640
+    },
+    {
+      "epoch": 1.280558789289872,
+      "grad_norm": 0.39471694827079773,
+      "learning_rate": 0.0002,
+      "loss": 1.7087,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2883197516492044,
+      "grad_norm": 0.35728853940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.7608,
+      "step": 1660
+    },
+    {
+      "epoch": 1.296080714008537,
+      "grad_norm": 0.35223081707954407,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3038416763678695,
+      "grad_norm": 0.3588867485523224,
+      "learning_rate": 0.0002,
+      "loss": 1.7253,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3116026387272022,
+      "grad_norm": 0.3528042733669281,
+      "learning_rate": 0.0002,
+      "loss": 1.6505,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3193636010865348,
+      "grad_norm": 0.35975801944732666,
+      "learning_rate": 0.0002,
+      "loss": 1.6945,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3271245634458673,
+      "grad_norm": 0.36691880226135254,
+      "learning_rate": 0.0002,
+      "loss": 1.6631,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3348855258052,
+      "grad_norm": 0.3787977695465088,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3426464881645324,
+      "grad_norm": 0.36614933609962463,
+      "learning_rate": 0.0002,
+      "loss": 1.7697,
+      "step": 1730
+    },
+    {
+      "epoch": 1.350407450523865,
+      "grad_norm": 0.3484745919704437,
+      "learning_rate": 0.0002,
+      "loss": 1.6487,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3581684128831975,
+      "grad_norm": 0.36905673146247864,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 1750
+    },
+    {
+      "epoch": 1.36592937524253,
+      "grad_norm": 0.41564738750457764,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3736903376018628,
+      "grad_norm": 0.3345205783843994,
+      "learning_rate": 0.0002,
+      "loss": 1.6634,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3814512999611952,
+      "grad_norm": 0.34926071763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7275,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3892122623205276,
+      "grad_norm": 0.42004233598709106,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3969732246798603,
+      "grad_norm": 0.3576236963272095,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 1800
+    },
+    {
+      "epoch": 1.404734187039193,
+      "grad_norm": 0.3586704432964325,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4124951493985254,
+      "grad_norm": 0.3943439722061157,
+      "learning_rate": 0.0002,
+      "loss": 1.6171,
+      "step": 1820
+    },
+    {
+      "epoch": 1.420256111757858,
+      "grad_norm": 0.3484877049922943,
+      "learning_rate": 0.0002,
+      "loss": 1.6865,
+      "step": 1830
+    },
+    {
+      "epoch": 1.4280170741171905,
+      "grad_norm": 0.3344518840312958,
+      "learning_rate": 0.0002,
+      "loss": 1.7205,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4357780364765231,
+      "grad_norm": 0.4345698356628418,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4435389988358556,
+      "grad_norm": 0.5525162220001221,
+      "learning_rate": 0.0002,
+      "loss": 1.6855,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4512999611951882,
+      "grad_norm": 0.37194496393203735,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1870
+    },
+    {
+      "epoch": 1.4590609235545209,
+      "grad_norm": 0.34570157527923584,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4668218859138533,
+      "grad_norm": 0.3512282073497772,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4745828482731858,
+      "grad_norm": 0.3443922996520996,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4823438106325184,
+      "grad_norm": 0.3812018036842346,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1910
+    },
+    {
+      "epoch": 1.490104772991851,
+      "grad_norm": 0.39263492822647095,
+      "learning_rate": 0.0002,
+      "loss": 1.7277,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4978657353511835,
+      "grad_norm": 0.3146156072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1930
+    },
+    {
+      "epoch": 1.505626697710516,
+      "grad_norm": 0.3653988540172577,
+      "learning_rate": 0.0002,
+      "loss": 1.6881,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5133876600698488,
+      "grad_norm": 0.3966596722602844,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5211486224291813,
+      "grad_norm": 0.3441697359085083,
+      "learning_rate": 0.0002,
+      "loss": 1.6942,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5289095847885137,
+      "grad_norm": 0.3328564465045929,
+      "learning_rate": 0.0002,
+      "loss": 1.7175,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5366705471478463,
+      "grad_norm": 0.34068772196769714,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 1980
+    },
+    {
+      "epoch": 1.544431509507179,
+      "grad_norm": 0.3559795916080475,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5521924718665114,
+      "grad_norm": 0.37888768315315247,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5599534342258439,
+      "grad_norm": 0.36128363013267517,
+      "learning_rate": 0.0002,
+      "loss": 1.7094,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5677143965851765,
+      "grad_norm": 0.3643714487552643,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5754753589445092,
+      "grad_norm": 0.3863612115383148,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5832363213038416,
+      "grad_norm": 0.32831457257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5909972836631743,
+      "grad_norm": 0.36098113656044006,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 2050
+    },
+    {
+      "epoch": 1.598758246022507,
+      "grad_norm": 1.1079334020614624,
+      "learning_rate": 0.0002,
+      "loss": 1.7065,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6065192083818394,
+      "grad_norm": 0.35615381598472595,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6142801707411718,
+      "grad_norm": 0.369711309671402,
+      "learning_rate": 0.0002,
+      "loss": 1.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6220411331005045,
+      "grad_norm": 0.390658438205719,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6298020954598371,
+      "grad_norm": 0.3422999382019043,
+      "learning_rate": 0.0002,
+      "loss": 1.6996,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6375630578191696,
+      "grad_norm": 0.372475266456604,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 2110
+    },
+    {
+      "epoch": 1.645324020178502,
+      "grad_norm": 0.35660576820373535,
+      "learning_rate": 0.0002,
+      "loss": 1.7216,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6530849825378346,
+      "grad_norm": 0.35754942893981934,
+      "learning_rate": 0.0002,
+      "loss": 1.6991,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6608459448971673,
+      "grad_norm": 0.34572410583496094,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6686069072564997,
+      "grad_norm": 0.42059701681137085,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6763678696158324,
+      "grad_norm": 0.35200759768486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2160
+    },
+    {
+      "epoch": 1.684128831975165,
+      "grad_norm": 0.3704029321670532,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6918897943344975,
+      "grad_norm": 0.40450501441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 2180
+    },
+    {
+      "epoch": 1.69965075669383,
+      "grad_norm": 0.362966924905777,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 2190
+    },
+    {
+      "epoch": 1.7074117190531626,
+      "grad_norm": 0.36586204171180725,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7151726814124952,
+      "grad_norm": 0.3295372426509857,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7229336437718277,
+      "grad_norm": 0.3892575800418854,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 2220
+    },
+    {
+      "epoch": 1.73069460613116,
+      "grad_norm": 0.34712135791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.7805,
+      "step": 2230
+    },
+    {
+      "epoch": 1.738455568490493,
+      "grad_norm": 0.34801796078681946,
+      "learning_rate": 0.0002,
+      "loss": 1.7353,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7462165308498254,
+      "grad_norm": 0.3822397291660309,
+      "learning_rate": 0.0002,
+      "loss": 1.7009,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7539774932091579,
+      "grad_norm": 0.38933250308036804,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7617384555684905,
+      "grad_norm": 0.3798373341560364,
+      "learning_rate": 0.0002,
+      "loss": 1.7245,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7694994179278232,
+      "grad_norm": 0.35151317715644836,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7772603802871556,
+      "grad_norm": 0.44981494545936584,
+      "learning_rate": 0.0002,
+      "loss": 1.6894,
+      "step": 2290
+    },
+    {
+      "epoch": 1.785021342646488,
+      "grad_norm": 0.3992624580860138,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7927823050058207,
+      "grad_norm": 0.3772512376308441,
+      "learning_rate": 0.0002,
+      "loss": 1.7252,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8005432673651534,
+      "grad_norm": 0.3511589467525482,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8083042297244858,
+      "grad_norm": 0.3805285394191742,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8160651920838184,
+      "grad_norm": 0.3792071044445038,
+      "learning_rate": 0.0002,
+      "loss": 1.6986,
+      "step": 2340
+    },
+    {
+      "epoch": 1.823826154443151,
+      "grad_norm": 0.36430829763412476,
+      "learning_rate": 0.0002,
+      "loss": 1.7759,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8315871168024835,
+      "grad_norm": 0.36502477526664734,
+      "learning_rate": 0.0002,
+      "loss": 1.6773,
+      "step": 2360
+    },
+    {
+      "epoch": 1.839348079161816,
+      "grad_norm": 0.35015153884887695,
+      "learning_rate": 0.0002,
+      "loss": 1.8072,
+      "step": 2370
+    },
+    {
+      "epoch": 1.8471090415211486,
+      "grad_norm": 0.3710903823375702,
+      "learning_rate": 0.0002,
+      "loss": 1.7734,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8548700038804813,
+      "grad_norm": 0.3542828857898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8626309662398137,
+      "grad_norm": 0.35467568039894104,
+      "learning_rate": 0.0002,
+      "loss": 1.6783,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8703919285991462,
+      "grad_norm": 0.3638560473918915,
+      "learning_rate": 0.0002,
+      "loss": 1.7773,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8781528909584788,
+      "grad_norm": 0.3823298215866089,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8859138533178115,
+      "grad_norm": 0.3926416337490082,
+      "learning_rate": 0.0002,
+      "loss": 1.6935,
+      "step": 2430
+    },
+    {
+      "epoch": 1.893674815677144,
+      "grad_norm": 0.3608079254627228,
+      "learning_rate": 0.0002,
+      "loss": 1.71,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9014357780364766,
+      "grad_norm": 0.3426613509654999,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9091967403958092,
+      "grad_norm": 0.3522338569164276,
+      "learning_rate": 0.0002,
+      "loss": 1.6892,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9169577027551417,
+      "grad_norm": 0.3608049154281616,
+      "learning_rate": 0.0002,
+      "loss": 1.7307,
+      "step": 2470
+    },
+    {
+      "epoch": 1.924718665114474,
+      "grad_norm": 0.3849755525588989,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9324796274738067,
+      "grad_norm": 0.4154011011123657,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9402405898331394,
+      "grad_norm": 0.3602796792984009,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9480015521924718,
+      "grad_norm": 0.3702992796897888,
+      "learning_rate": 0.0002,
+      "loss": 1.7843,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9557625145518043,
+      "grad_norm": 0.3657735288143158,
+      "learning_rate": 0.0002,
+      "loss": 1.6669,
+      "step": 2520
+    },
+    {
+      "epoch": 1.963523476911137,
+      "grad_norm": 0.41031739115715027,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9712844392704696,
+      "grad_norm": 0.34578680992126465,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 2540
+    },
+    {
+      "epoch": 1.979045401629802,
+      "grad_norm": 0.3361521065235138,
+      "learning_rate": 0.0002,
+      "loss": 1.723,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9868063639891347,
+      "grad_norm": 0.34342363476753235,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9945673263484673,
+      "grad_norm": 0.32954007387161255,
+      "learning_rate": 0.0002,
+      "loss": 1.6577,
+      "step": 2570
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8068748712539673,
+      "eval_runtime": 105.5885,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 2577
+    },
+    {
+      "epoch": 2.0023282887077998,
+      "grad_norm": 0.336302250623703,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 2580
+    },
+    {
+      "epoch": 2.010089251067132,
+      "grad_norm": 0.3627048432826996,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2590
+    },
+    {
+      "epoch": 2.017850213426465,
+      "grad_norm": 0.38406702876091003,
+      "learning_rate": 0.0002,
+      "loss": 1.4908,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0256111757857975,
+      "grad_norm": 0.5326781272888184,
+      "learning_rate": 0.0002,
+      "loss": 1.5368,
+      "step": 2610
+    },
+    {
+      "epoch": 2.03337213814513,
+      "grad_norm": 0.4774554967880249,
+      "learning_rate": 0.0002,
+      "loss": 1.5727,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0411331005044624,
+      "grad_norm": 0.4251810312271118,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 2630
+    },
+    {
+      "epoch": 2.0488940628637953,
+      "grad_norm": 0.4693007171154022,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 2640
+    },
+    {
+      "epoch": 2.0566550252231277,
+      "grad_norm": 0.46371519565582275,
+      "learning_rate": 0.0002,
+      "loss": 1.6137,
+      "step": 2650
+    },
+    {
+      "epoch": 2.06441598758246,
+      "grad_norm": 0.46652570366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0721769499417926,
+      "grad_norm": 0.45200315117836,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0799379123011255,
+      "grad_norm": 0.42905205488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 2680
+    },
+    {
+      "epoch": 2.087698874660458,
+      "grad_norm": 0.44509148597717285,
+      "learning_rate": 0.0002,
+      "loss": 1.5401,
+      "step": 2690
+    },
+    {
+      "epoch": 2.0954598370197903,
+      "grad_norm": 0.4445319175720215,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 2700
+    },
+    {
+      "epoch": 2.103220799379123,
+      "grad_norm": 0.46825504302978516,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1109817617384556,
+      "grad_norm": 0.4623856842517853,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 2720
+    },
+    {
+      "epoch": 2.118742724097788,
+      "grad_norm": 0.4833452105522156,
+      "learning_rate": 0.0002,
+      "loss": 1.5601,
+      "step": 2730
+    },
+    {
+      "epoch": 2.1265036864571205,
+      "grad_norm": 0.4582686722278595,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2740
+    },
+    {
+      "epoch": 2.1342646488164534,
+      "grad_norm": 0.47587934136390686,
+      "learning_rate": 0.0002,
+      "loss": 1.5801,
+      "step": 2750
+    },
+    {
+      "epoch": 2.142025611175786,
+      "grad_norm": 0.4602217972278595,
+      "learning_rate": 0.0002,
+      "loss": 1.594,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1497865735351183,
+      "grad_norm": 0.47501352429389954,
+      "learning_rate": 0.0002,
+      "loss": 1.5271,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1575475358944507,
+      "grad_norm": 0.5078499913215637,
+      "learning_rate": 0.0002,
+      "loss": 1.4862,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1653084982537836,
+      "grad_norm": 0.497704416513443,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 2790
+    },
+    {
+      "epoch": 2.173069460613116,
+      "grad_norm": 0.5435971617698669,
+      "learning_rate": 0.0002,
+      "loss": 1.5597,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1808304229724484,
+      "grad_norm": 0.5172356367111206,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1885913853317813,
+      "grad_norm": 0.44063422083854675,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 2820
+    },
+    {
+      "epoch": 2.1963523476911138,
+      "grad_norm": 0.5079569220542908,
+      "learning_rate": 0.0002,
+      "loss": 1.6041,
+      "step": 2830
+    },
+    {
+      "epoch": 2.204113310050446,
+      "grad_norm": 0.45658132433891296,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2840
+    },
+    {
+      "epoch": 2.2118742724097786,
+      "grad_norm": 0.5103023648262024,
+      "learning_rate": 0.0002,
+      "loss": 1.5546,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2196352347691115,
+      "grad_norm": 0.4882226288318634,
+      "learning_rate": 0.0002,
+      "loss": 1.6197,
+      "step": 2860
+    },
+    {
+      "epoch": 2.227396197128444,
+      "grad_norm": 0.5087296962738037,
+      "learning_rate": 0.0002,
+      "loss": 1.5996,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2351571594877764,
+      "grad_norm": 0.45293712615966797,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 2880
+    },
+    {
+      "epoch": 2.242918121847109,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.0002,
+      "loss": 1.6214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2506790842064417,
+      "grad_norm": 0.47126415371894836,
+      "learning_rate": 0.0002,
+      "loss": 1.5273,
+      "step": 2900
+    },
+    {
+      "epoch": 2.258440046565774,
+      "grad_norm": 0.44005846977233887,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 2910
+    },
+    {
+      "epoch": 2.2662010089251066,
+      "grad_norm": 0.46476176381111145,
+      "learning_rate": 0.0002,
+      "loss": 1.6023,
+      "step": 2920
+    },
+    {
+      "epoch": 2.2739619712844394,
+      "grad_norm": 0.48051515221595764,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 2930
+    },
+    {
+      "epoch": 2.281722933643772,
+      "grad_norm": 0.480069637298584,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2894838960031043,
+      "grad_norm": 0.5122102499008179,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 2950
+    },
+    {
+      "epoch": 2.2972448583624367,
+      "grad_norm": 0.48879891633987427,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3050058207217696,
+      "grad_norm": 0.4973136782646179,
+      "learning_rate": 0.0002,
+      "loss": 1.5483,
+      "step": 2970
+    },
+    {
+      "epoch": 2.312766783081102,
+      "grad_norm": 0.5522695183753967,
+      "learning_rate": 0.0002,
+      "loss": 1.677,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3205277454404345,
+      "grad_norm": 0.5220217704772949,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 2990
+    },
+    {
+      "epoch": 2.328288707799767,
+      "grad_norm": 0.4978662431240082,
+      "learning_rate": 0.0002,
+      "loss": 1.6299,
+      "step": 3000
+    },
+    {
+      "epoch": 2.3360496701591,
+      "grad_norm": 0.554053544998169,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 3010
+    },
+    {
+      "epoch": 2.3438106325184322,
+      "grad_norm": 0.4703886806964874,
+      "learning_rate": 0.0002,
+      "loss": 1.5356,
+      "step": 3020
+    },
+    {
+      "epoch": 2.3515715948777647,
+      "grad_norm": 0.5074123740196228,
+      "learning_rate": 0.0002,
+      "loss": 1.5418,
+      "step": 3030
+    },
+    {
+      "epoch": 2.3593325572370976,
+      "grad_norm": 0.5088278651237488,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 3040
+    },
+    {
+      "epoch": 2.36709351959643,
+      "grad_norm": 0.4752114415168762,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3748544819557624,
+      "grad_norm": 0.5121659636497498,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3826154443150953,
+      "grad_norm": 0.48649218678474426,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3070
+    },
+    {
+      "epoch": 2.3903764066744277,
+      "grad_norm": 0.5209488868713379,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 3080
+    },
+    {
+      "epoch": 2.39813736903376,
+      "grad_norm": 0.5110517740249634,
+      "learning_rate": 0.0002,
+      "loss": 1.597,
+      "step": 3090
+    },
+    {
+      "epoch": 2.4058983313930926,
+      "grad_norm": 0.5609337091445923,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4136592937524255,
+      "grad_norm": 0.5191826224327087,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 3110
+    },
+    {
+      "epoch": 2.421420256111758,
+      "grad_norm": 0.4876069724559784,
+      "learning_rate": 0.0002,
+      "loss": 1.6347,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4291812184710904,
+      "grad_norm": 0.4713933765888214,
+      "learning_rate": 0.0002,
+      "loss": 1.5565,
+      "step": 3130
+    },
+    {
+      "epoch": 2.436942180830423,
+      "grad_norm": 0.5102227330207825,
+      "learning_rate": 0.0002,
+      "loss": 1.6388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4447031431897557,
+      "grad_norm": 0.44546666741371155,
+      "learning_rate": 0.0002,
+      "loss": 1.5667,
+      "step": 3150
+    },
+    {
+      "epoch": 2.452464105549088,
+      "grad_norm": 0.5167558193206787,
+      "learning_rate": 0.0002,
+      "loss": 1.5973,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4602250679084205,
+      "grad_norm": 0.5226958990097046,
+      "learning_rate": 0.0002,
+      "loss": 1.5673,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4679860302677534,
+      "grad_norm": 0.4751799702644348,
+      "learning_rate": 0.0002,
+      "loss": 1.5758,
+      "step": 3180
+    },
+    {
+      "epoch": 2.475746992627086,
+      "grad_norm": 0.4744729697704315,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.4835079549864183,
+      "grad_norm": 0.5203230381011963,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 3200
+    },
+    {
+      "epoch": 2.4912689173457507,
+      "grad_norm": 0.47209781408309937,
+      "learning_rate": 0.0002,
+      "loss": 1.493,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4990298797050836,
+      "grad_norm": 0.5241674780845642,
+      "learning_rate": 0.0002,
+      "loss": 1.6415,
+      "step": 3220
+    },
+    {
+      "epoch": 2.506790842064416,
+      "grad_norm": 0.5152244567871094,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5145518044237485,
+      "grad_norm": 0.5216741561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.6248,
+      "step": 3240
+    },
+    {
+      "epoch": 2.522312766783081,
+      "grad_norm": 0.4953259527683258,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 3250
+    },
+    {
+      "epoch": 2.530073729142414,
+      "grad_norm": 0.5973829030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.666,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5378346915017462,
+      "grad_norm": 0.48804202675819397,
+      "learning_rate": 0.0002,
+      "loss": 1.5295,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5455956538610787,
+      "grad_norm": 0.5334644317626953,
+      "learning_rate": 0.0002,
+      "loss": 1.4954,
+      "step": 3280
+    },
+    {
+      "epoch": 2.5533566162204115,
+      "grad_norm": 0.46873313188552856,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3290
+    },
+    {
+      "epoch": 2.561117578579744,
+      "grad_norm": 0.4282589554786682,
+      "learning_rate": 0.0002,
+      "loss": 1.5362,
+      "step": 3300
+    },
+    {
+      "epoch": 2.5688785409390764,
+      "grad_norm": 0.4848293960094452,
+      "learning_rate": 0.0002,
+      "loss": 1.6278,
+      "step": 3310
+    },
+    {
+      "epoch": 2.576639503298409,
+      "grad_norm": 0.5093745589256287,
+      "learning_rate": 0.0002,
+      "loss": 1.6308,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5844004656577413,
+      "grad_norm": 0.5084842443466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 3330
+    },
+    {
+      "epoch": 2.592161428017074,
+      "grad_norm": 0.4696281850337982,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3340
+    },
+    {
+      "epoch": 2.5999223903764066,
+      "grad_norm": 0.5767765641212463,
+      "learning_rate": 0.0002,
+      "loss": 1.5359,
+      "step": 3350
+    },
+    {
+      "epoch": 2.607683352735739,
+      "grad_norm": 0.47300875186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.6097,
+      "step": 3360
+    },
+    {
+      "epoch": 2.615444315095072,
+      "grad_norm": 0.4809158146381378,
+      "learning_rate": 0.0002,
+      "loss": 1.6138,
+      "step": 3370
+    },
+    {
+      "epoch": 2.6232052774544043,
+      "grad_norm": 0.5141063928604126,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 3380
+    },
+    {
+      "epoch": 2.630966239813737,
+      "grad_norm": 0.4832935035228729,
+      "learning_rate": 0.0002,
+      "loss": 1.5784,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6387272021730697,
+      "grad_norm": 0.5044625401496887,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3400
+    },
+    {
+      "epoch": 2.646488164532402,
+      "grad_norm": 0.5287680625915527,
+      "learning_rate": 0.0002,
+      "loss": 1.6202,
+      "step": 3410
+    },
+    {
+      "epoch": 2.6542491268917345,
+      "grad_norm": 0.5306379795074463,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 3420
+    },
+    {
+      "epoch": 2.662010089251067,
+      "grad_norm": 0.5849291682243347,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6697710516104,
+      "grad_norm": 0.7951080799102783,
+      "learning_rate": 0.0002,
+      "loss": 1.5937,
+      "step": 3440
+    },
+    {
+      "epoch": 2.6775320139697323,
+      "grad_norm": 0.48087653517723083,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6852929763290647,
+      "grad_norm": 0.5396431684494019,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 3460
+    },
+    {
+      "epoch": 2.693053938688397,
+      "grad_norm": 0.5481634736061096,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 3470
+    },
+    {
+      "epoch": 2.70081490104773,
+      "grad_norm": 0.5068731307983398,
+      "learning_rate": 0.0002,
+      "loss": 1.6436,
+      "step": 3480
+    },
+    {
+      "epoch": 2.7085758634070625,
+      "grad_norm": 0.5759826898574829,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 3490
+    },
+    {
+      "epoch": 2.716336825766395,
+      "grad_norm": 0.7253932952880859,
+      "learning_rate": 0.0002,
+      "loss": 1.596,
+      "step": 3500
+    },
+    {
+      "epoch": 2.724097788125728,
+      "grad_norm": 0.527745246887207,
+      "learning_rate": 0.0002,
+      "loss": 1.5791,
+      "step": 3510
+    },
+    {
+      "epoch": 2.73185875048506,
+      "grad_norm": 0.5279242396354675,
+      "learning_rate": 0.0002,
+      "loss": 1.5874,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7396197128443927,
+      "grad_norm": 0.5047839283943176,
+      "learning_rate": 0.0002,
+      "loss": 1.6768,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7473806752037255,
+      "grad_norm": 0.5430883169174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5517,
+      "step": 3540
+    },
+    {
+      "epoch": 2.755141637563058,
+      "grad_norm": 0.4496723711490631,
+      "learning_rate": 0.0002,
+      "loss": 1.5624,
+      "step": 3550
+    },
+    {
+      "epoch": 2.7629025999223904,
+      "grad_norm": 0.5063338875770569,
+      "learning_rate": 0.0002,
+      "loss": 1.5789,
+      "step": 3560
+    },
+    {
+      "epoch": 2.770663562281723,
+      "grad_norm": 0.4619026780128479,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 3570
+    },
+    {
+      "epoch": 2.7784245246410553,
+      "grad_norm": 0.4753304123878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5793,
+      "step": 3580
+    },
+    {
+      "epoch": 2.786185487000388,
+      "grad_norm": 0.5422708988189697,
+      "learning_rate": 0.0002,
+      "loss": 1.5715,
+      "step": 3590
+    },
+    {
+      "epoch": 2.7939464493597206,
+      "grad_norm": 0.4756578803062439,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 3600
+    },
+    {
+      "epoch": 2.801707411719053,
+      "grad_norm": 0.5057567358016968,
+      "learning_rate": 0.0002,
+      "loss": 1.5358,
+      "step": 3610
+    },
+    {
+      "epoch": 2.809468374078386,
+      "grad_norm": 0.5410919785499573,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8172293364377183,
+      "grad_norm": 0.4958136975765228,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8249902987970508,
+      "grad_norm": 0.454527348279953,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 3640
+    },
+    {
+      "epoch": 2.8327512611563836,
+      "grad_norm": 0.5092706084251404,
+      "learning_rate": 0.0002,
+      "loss": 1.5582,
+      "step": 3650
+    },
+    {
+      "epoch": 2.840512223515716,
+      "grad_norm": 0.5314022302627563,
+      "learning_rate": 0.0002,
+      "loss": 1.5893,
+      "step": 3660
+    },
+    {
+      "epoch": 2.8482731858750485,
+      "grad_norm": 0.5028239488601685,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3670
+    },
+    {
+      "epoch": 2.856034148234381,
+      "grad_norm": 0.5127444863319397,
+      "learning_rate": 0.0002,
+      "loss": 1.5751,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8637951105937134,
+      "grad_norm": 0.5045645236968994,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8715560729530463,
+      "grad_norm": 0.5560781955718994,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8793170353123787,
+      "grad_norm": 0.5177600383758545,
+      "learning_rate": 0.0002,
+      "loss": 1.5988,
+      "step": 3710
+    },
+    {
+      "epoch": 2.887077997671711,
+      "grad_norm": 0.45830899477005005,
+      "learning_rate": 0.0002,
+      "loss": 1.6009,
+      "step": 3720
+    },
+    {
+      "epoch": 2.894838960031044,
+      "grad_norm": 0.4828629195690155,
+      "learning_rate": 0.0002,
+      "loss": 1.6344,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9025999223903765,
+      "grad_norm": 0.48241183161735535,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3740
+    },
+    {
+      "epoch": 2.910360884749709,
+      "grad_norm": 0.4909592568874359,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 3750
+    },
+    {
+      "epoch": 2.9181218471090418,
+      "grad_norm": 0.44677025079727173,
+      "learning_rate": 0.0002,
+      "loss": 1.4927,
+      "step": 3760
+    },
+    {
+      "epoch": 2.925882809468374,
+      "grad_norm": 0.4928834140300751,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 3770
+    },
+    {
+      "epoch": 2.9336437718277066,
+      "grad_norm": 0.5673553347587585,
+      "learning_rate": 0.0002,
+      "loss": 1.5843,
+      "step": 3780
+    },
+    {
+      "epoch": 2.941404734187039,
+      "grad_norm": 0.548190712928772,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9491656965463715,
+      "grad_norm": 0.48979803919792175,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 3800
+    },
+    {
+      "epoch": 2.9569266589057044,
+      "grad_norm": 0.533191978931427,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 3810
+    },
+    {
+      "epoch": 2.964687621265037,
+      "grad_norm": 0.5362946391105652,
+      "learning_rate": 0.0002,
+      "loss": 1.584,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9724485836243693,
+      "grad_norm": 0.4724906384944916,
+      "learning_rate": 0.0002,
+      "loss": 1.6602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.980209545983702,
+      "grad_norm": 0.5468461513519287,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 3840
+    },
+    {
+      "epoch": 2.9879705083430346,
+      "grad_norm": 0.4697108864784241,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 3850
+    },
+    {
+      "epoch": 2.995731470702367,
+      "grad_norm": 0.4780906140804291,
+      "learning_rate": 0.0002,
+      "loss": 1.6312,
+      "step": 3860
+    },
+    {
+      "epoch": 2.9996119518820334,
+      "eval_loss": 1.8472607135772705,
+      "eval_runtime": 106.5541,
+      "eval_samples_per_second": 4.758,
+      "eval_steps_per_second": 0.601,
+      "step": 3865
+    },
+    {
+      "epoch": 3.0034924330616994,
+      "grad_norm": 0.5645653605461121,
+      "learning_rate": 0.0002,
+      "loss": 1.4983,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0112533954210323,
+      "grad_norm": 0.6457151174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0190143577803648,
+      "grad_norm": 0.583838164806366,
+      "learning_rate": 0.0002,
+      "loss": 1.3899,
+      "step": 3890
+    },
+    {
+      "epoch": 3.026775320139697,
+      "grad_norm": 0.6819260120391846,
+      "learning_rate": 0.0002,
+      "loss": 1.3258,
+      "step": 3900
+    },
+    {
+      "epoch": 3.03453628249903,
+      "grad_norm": 0.6692903637886047,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0422972448583625,
+      "grad_norm": 0.6101024746894836,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 3920
+    },
+    {
+      "epoch": 3.050058207217695,
+      "grad_norm": 0.7014093399047852,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3930
+    },
+    {
+      "epoch": 3.0578191695770274,
+      "grad_norm": 0.7380381226539612,
+      "learning_rate": 0.0002,
+      "loss": 1.3885,
+      "step": 3940
+    },
+    {
+      "epoch": 3.0655801319363603,
+      "grad_norm": 0.6607900857925415,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0733410942956927,
+      "grad_norm": 0.735263466835022,
+      "learning_rate": 0.0002,
+      "loss": 1.4293,
+      "step": 3960
+    },
+    {
+      "epoch": 3.081102056655025,
+      "grad_norm": 0.6788513660430908,
+      "learning_rate": 0.0002,
+      "loss": 1.3966,
+      "step": 3970
+    },
+    {
+      "epoch": 3.088863019014358,
+      "grad_norm": 0.6347652673721313,
+      "learning_rate": 0.0002,
+      "loss": 1.3435,
+      "step": 3980
+    },
+    {
+      "epoch": 3.0966239813736904,
+      "grad_norm": 0.7056642770767212,
+      "learning_rate": 0.0002,
+      "loss": 1.4518,
+      "step": 3990
+    },
+    {
+      "epoch": 3.104384943733023,
+      "grad_norm": 0.6387075185775757,
+      "learning_rate": 0.0002,
+      "loss": 1.4474,
+      "step": 4000
+    },
+    {
+      "epoch": 3.1121459060923553,
+      "grad_norm": 0.6701116561889648,
+      "learning_rate": 0.0002,
+      "loss": 1.3833,
+      "step": 4010
+    },
+    {
+      "epoch": 3.119906868451688,
+      "grad_norm": 0.7558449506759644,
+      "learning_rate": 0.0002,
+      "loss": 1.404,
+      "step": 4020
+    },
+    {
+      "epoch": 3.1276678308110206,
+      "grad_norm": 0.6612881422042847,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 4030
+    },
+    {
+      "epoch": 3.135428793170353,
+      "grad_norm": 0.7474587559700012,
+      "learning_rate": 0.0002,
+      "loss": 1.439,
+      "step": 4040
+    },
+    {
+      "epoch": 3.1431897555296855,
+      "grad_norm": 0.7292373776435852,
+      "learning_rate": 0.0002,
+      "loss": 1.4616,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1509507178890184,
+      "grad_norm": 0.7432886958122253,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 4060
+    },
+    {
+      "epoch": 3.158711680248351,
+      "grad_norm": 0.6366098523139954,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1664726426076832,
+      "grad_norm": 0.6837611794471741,
+      "learning_rate": 0.0002,
+      "loss": 1.5044,
+      "step": 4080
+    },
+    {
+      "epoch": 3.174233604967016,
+      "grad_norm": 0.7194393277168274,
+      "learning_rate": 0.0002,
+      "loss": 1.4332,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1819945673263486,
+      "grad_norm": 0.6963607668876648,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 4100
+    },
+    {
+      "epoch": 3.189755529685681,
+      "grad_norm": 0.6404902935028076,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4110
+    },
+    {
+      "epoch": 3.1975164920450134,
+      "grad_norm": 0.7172070741653442,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 4120
+    },
+    {
+      "epoch": 3.2052774544043463,
+      "grad_norm": 0.6577759385108948,
+      "learning_rate": 0.0002,
+      "loss": 1.4658,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2130384167636787,
+      "grad_norm": 0.6658480167388916,
+      "learning_rate": 0.0002,
+      "loss": 1.4019,
+      "step": 4140
+    },
+    {
+      "epoch": 3.220799379123011,
+      "grad_norm": 0.6771699786186218,
+      "learning_rate": 0.0002,
+      "loss": 1.4348,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2285603414823436,
+      "grad_norm": 0.699035108089447,
+      "learning_rate": 0.0002,
+      "loss": 1.4736,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2363213038416765,
+      "grad_norm": 0.7218514680862427,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 4170
+    },
+    {
+      "epoch": 3.244082266201009,
+      "grad_norm": 0.6270631551742554,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 4180
+    },
+    {
+      "epoch": 3.2518432285603414,
+      "grad_norm": 0.6828921437263489,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2596041909196742,
+      "grad_norm": 0.6005498170852661,
+      "learning_rate": 0.0002,
+      "loss": 1.4663,
+      "step": 4200
+    },
+    {
+      "epoch": 3.2673651532790067,
+      "grad_norm": 0.6974790692329407,
+      "learning_rate": 0.0002,
+      "loss": 1.4798,
+      "step": 4210
+    },
+    {
+      "epoch": 3.275126115638339,
+      "grad_norm": 0.7269543409347534,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4220
+    },
+    {
+      "epoch": 3.2828870779976715,
+      "grad_norm": 0.6728787422180176,
+      "learning_rate": 0.0002,
+      "loss": 1.3848,
+      "step": 4230
+    },
+    {
+      "epoch": 3.2906480403570044,
+      "grad_norm": 0.676972508430481,
+      "learning_rate": 0.0002,
+      "loss": 1.4112,
+      "step": 4240
+    },
+    {
+      "epoch": 3.298409002716337,
+      "grad_norm": 0.748309314250946,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3061699650756693,
+      "grad_norm": 0.6976589560508728,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 4260
+    },
+    {
+      "epoch": 3.3139309274350017,
+      "grad_norm": 0.649780809879303,
+      "learning_rate": 0.0002,
+      "loss": 1.3967,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3216918897943346,
+      "grad_norm": 0.6529902815818787,
+      "learning_rate": 0.0002,
+      "loss": 1.327,
+      "step": 4280
+    },
+    {
+      "epoch": 3.329452852153667,
+      "grad_norm": 0.9273163676261902,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 4290
+    },
+    {
+      "epoch": 3.3372138145129995,
+      "grad_norm": 0.717024028301239,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 4300
+    },
+    {
+      "epoch": 3.3449747768723324,
+      "grad_norm": 0.7914950251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.4441,
+      "step": 4310
+    },
+    {
+      "epoch": 3.352735739231665,
+      "grad_norm": 0.7133203148841858,
+      "learning_rate": 0.0002,
+      "loss": 1.432,
+      "step": 4320
+    },
+    {
+      "epoch": 3.3604967015909972,
+      "grad_norm": 0.7409568428993225,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3682576639503297,
+      "grad_norm": 0.6993981003761292,
+      "learning_rate": 0.0002,
+      "loss": 1.3992,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3760186263096625,
+      "grad_norm": 0.7114535570144653,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4350
+    },
+    {
+      "epoch": 3.383779588668995,
+      "grad_norm": 0.6790860295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.4227,
+      "step": 4360
+    },
+    {
+      "epoch": 3.3915405510283274,
+      "grad_norm": 0.6507849097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 4370
+    },
+    {
+      "epoch": 3.39930151338766,
+      "grad_norm": 0.5967804193496704,
+      "learning_rate": 0.0002,
+      "loss": 1.4559,
+      "step": 4380
+    },
+    {
+      "epoch": 3.4070624757469927,
+      "grad_norm": 0.6625847816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 4390
+    },
+    {
+      "epoch": 3.414823438106325,
+      "grad_norm": 0.6736508011817932,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4225844004656576,
+      "grad_norm": 0.7870860695838928,
+      "learning_rate": 0.0002,
+      "loss": 1.4363,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4303453628249905,
+      "grad_norm": 0.7205295562744141,
+      "learning_rate": 0.0002,
+      "loss": 1.4114,
+      "step": 4420
+    },
+    {
+      "epoch": 3.438106325184323,
+      "grad_norm": 0.6634634137153625,
+      "learning_rate": 0.0002,
+      "loss": 1.4131,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4458672875436553,
+      "grad_norm": 0.7562733292579651,
+      "learning_rate": 0.0002,
+      "loss": 1.4683,
+      "step": 4440
+    },
+    {
+      "epoch": 3.453628249902988,
+      "grad_norm": 0.6585879921913147,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4613892122623207,
+      "grad_norm": 0.6896792054176331,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 4460
+    },
+    {
+      "epoch": 3.469150174621653,
+      "grad_norm": 0.6520342230796814,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4769111369809855,
+      "grad_norm": 0.6760806441307068,
+      "learning_rate": 0.0002,
+      "loss": 1.3423,
+      "step": 4480
+    },
+    {
+      "epoch": 3.484672099340318,
+      "grad_norm": 0.7539774179458618,
+      "learning_rate": 0.0002,
+      "loss": 1.4398,
+      "step": 4490
+    },
+    {
+      "epoch": 3.492433061699651,
+      "grad_norm": 0.7409411668777466,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 4500
+    },
+    {
+      "epoch": 3.5001940240589833,
+      "grad_norm": 0.6876253485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.4069,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5079549864183157,
+      "grad_norm": 0.7028461694717407,
+      "learning_rate": 0.0002,
+      "loss": 1.4228,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5157159487776486,
+      "grad_norm": 0.8056529760360718,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4530
+    },
+    {
+      "epoch": 3.523476911136981,
+      "grad_norm": 0.711338996887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5312378734963135,
+      "grad_norm": 0.7343552708625793,
+      "learning_rate": 0.0002,
+      "loss": 1.5247,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5389988358556463,
+      "grad_norm": 0.745479941368103,
+      "learning_rate": 0.0002,
+      "loss": 1.4308,
+      "step": 4560
+    },
+    {
+      "epoch": 3.5467597982149788,
+      "grad_norm": 0.7582294940948486,
+      "learning_rate": 0.0002,
+      "loss": 1.4229,
+      "step": 4570
+    },
+    {
+      "epoch": 3.554520760574311,
+      "grad_norm": 0.6717444658279419,
+      "learning_rate": 0.0002,
+      "loss": 1.4127,
+      "step": 4580
+    },
+    {
+      "epoch": 3.5622817229336436,
+      "grad_norm": 0.7417883276939392,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 4590
+    },
+    {
+      "epoch": 3.570042685292976,
+      "grad_norm": 0.6385737061500549,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4600
+    },
+    {
+      "epoch": 3.577803647652309,
+      "grad_norm": 0.716704249382019,
+      "learning_rate": 0.0002,
+      "loss": 1.3981,
+      "step": 4610
+    },
+    {
+      "epoch": 3.5855646100116414,
+      "grad_norm": 0.6948980093002319,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 4620
+    },
+    {
+      "epoch": 3.593325572370974,
+      "grad_norm": 0.6961140036582947,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6010865347303067,
+      "grad_norm": 0.7493122220039368,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 4640
+    },
+    {
+      "epoch": 3.608847497089639,
+      "grad_norm": 0.7431658506393433,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6166084594489716,
+      "grad_norm": 0.8353387713432312,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4660
+    },
+    {
+      "epoch": 3.6243694218083045,
+      "grad_norm": 0.7095612287521362,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 4670
+    },
+    {
+      "epoch": 3.632130384167637,
+      "grad_norm": 0.776620090007782,
+      "learning_rate": 0.0002,
+      "loss": 1.4003,
+      "step": 4680
+    },
+    {
+      "epoch": 3.6398913465269693,
+      "grad_norm": 0.7198925018310547,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6476523088863018,
+      "grad_norm": 0.8238834738731384,
+      "learning_rate": 0.0002,
+      "loss": 1.4543,
+      "step": 4700
+    },
+    {
+      "epoch": 3.655413271245634,
+      "grad_norm": 0.6804245710372925,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4710
+    },
+    {
+      "epoch": 3.663174233604967,
+      "grad_norm": 0.8444845676422119,
+      "learning_rate": 0.0002,
+      "loss": 1.4158,
+      "step": 4720
+    },
+    {
+      "epoch": 3.6709351959642995,
+      "grad_norm": 0.743797779083252,
+      "learning_rate": 0.0002,
+      "loss": 1.3825,
+      "step": 4730
+    },
+    {
+      "epoch": 3.678696158323632,
+      "grad_norm": 0.8994188904762268,
+      "learning_rate": 0.0002,
+      "loss": 1.4213,
+      "step": 4740
+    },
+    {
+      "epoch": 3.686457120682965,
+      "grad_norm": 0.75416100025177,
+      "learning_rate": 0.0002,
+      "loss": 1.4281,
+      "step": 4750
+    },
+    {
+      "epoch": 3.6942180830422973,
+      "grad_norm": 0.6499266028404236,
+      "learning_rate": 0.0002,
+      "loss": 1.4154,
+      "step": 4760
+    },
+    {
+      "epoch": 3.7019790454016297,
+      "grad_norm": 0.7246791124343872,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7097400077609626,
+      "grad_norm": 0.7831124067306519,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 4780
+    },
+    {
+      "epoch": 3.717500970120295,
+      "grad_norm": 0.7130028009414673,
+      "learning_rate": 0.0002,
+      "loss": 1.3933,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7252619324796274,
+      "grad_norm": 0.7501602172851562,
+      "learning_rate": 0.0002,
+      "loss": 1.4632,
+      "step": 4800
+    },
+    {
+      "epoch": 3.73302289483896,
+      "grad_norm": 0.6980932950973511,
+      "learning_rate": 0.0002,
+      "loss": 1.4985,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7407838571982923,
+      "grad_norm": 0.8050530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.4517,
+      "step": 4820
+    },
+    {
+      "epoch": 3.748544819557625,
+      "grad_norm": 0.6385579705238342,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7563057819169576,
+      "grad_norm": 0.6664714813232422,
+      "learning_rate": 0.0002,
+      "loss": 1.5281,
+      "step": 4840
+    },
+    {
+      "epoch": 3.76406674427629,
+      "grad_norm": 0.7125676274299622,
+      "learning_rate": 0.0002,
+      "loss": 1.4443,
+      "step": 4850
+    },
+    {
+      "epoch": 3.771827706635623,
+      "grad_norm": 0.7231866717338562,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 4860
+    },
+    {
+      "epoch": 3.7795886689949554,
+      "grad_norm": 0.6917183995246887,
+      "learning_rate": 0.0002,
+      "loss": 1.4446,
+      "step": 4870
+    },
+    {
+      "epoch": 3.787349631354288,
+      "grad_norm": 0.665037989616394,
+      "learning_rate": 0.0002,
+      "loss": 1.4369,
+      "step": 4880
+    },
+    {
+      "epoch": 3.7951105937136207,
+      "grad_norm": 0.5837726593017578,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 4890
+    },
+    {
+      "epoch": 3.802871556072953,
+      "grad_norm": 0.6366701722145081,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8106325184322856,
+      "grad_norm": 0.7082223892211914,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 4910
+    },
+    {
+      "epoch": 3.818393480791618,
+      "grad_norm": 0.8101672530174255,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4920
+    },
+    {
+      "epoch": 3.826154443150951,
+      "grad_norm": 0.7516148090362549,
+      "learning_rate": 0.0002,
+      "loss": 1.3659,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8339154055102833,
+      "grad_norm": 0.7928489446640015,
+      "learning_rate": 0.0002,
+      "loss": 1.3909,
+      "step": 4940
+    },
+    {
+      "epoch": 3.8416763678696157,
+      "grad_norm": 0.6892234683036804,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 4950
+    },
+    {
+      "epoch": 3.849437330228948,
+      "grad_norm": 0.6381304264068604,
+      "learning_rate": 0.0002,
+      "loss": 1.5024,
+      "step": 4960
+    },
+    {
+      "epoch": 3.857198292588281,
+      "grad_norm": 0.8068831562995911,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8649592549476135,
+      "grad_norm": 0.7289869785308838,
+      "learning_rate": 0.0002,
+      "loss": 1.45,
+      "step": 4980
+    },
+    {
+      "epoch": 3.872720217306946,
+      "grad_norm": 0.7278549075126648,
+      "learning_rate": 0.0002,
+      "loss": 1.398,
+      "step": 4990
+    },
+    {
+      "epoch": 3.880481179666279,
+      "grad_norm": 0.7324236631393433,
+      "learning_rate": 0.0002,
+      "loss": 1.4442,
+      "step": 5000
+    },
+    {
+      "epoch": 3.8882421420256112,
+      "grad_norm": 0.6759871244430542,
+      "learning_rate": 0.0002,
+      "loss": 1.4511,
+      "step": 5010
+    },
+    {
+      "epoch": 3.8960031043849437,
+      "grad_norm": 0.8159207701683044,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 5020
+    },
+    {
+      "epoch": 3.9037640667442766,
+      "grad_norm": 0.6536211967468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5030
+    },
+    {
+      "epoch": 3.911525029103609,
+      "grad_norm": 0.6827932000160217,
+      "learning_rate": 0.0002,
+      "loss": 1.4335,
+      "step": 5040
+    },
+    {
+      "epoch": 3.9192859914629414,
+      "grad_norm": 0.6688340306282043,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 5050
+    },
+    {
+      "epoch": 3.927046953822274,
+      "grad_norm": 0.6385695934295654,
+      "learning_rate": 0.0002,
+      "loss": 1.4099,
+      "step": 5060
+    },
+    {
+      "epoch": 3.9348079161816063,
+      "grad_norm": 0.6975107192993164,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 5070
+    },
+    {
+      "epoch": 3.942568878540939,
+      "grad_norm": 0.6684112548828125,
+      "learning_rate": 0.0002,
+      "loss": 1.4893,
+      "step": 5080
+    },
+    {
+      "epoch": 3.9503298409002716,
+      "grad_norm": 0.8349628448486328,
+      "learning_rate": 0.0002,
+      "loss": 1.4732,
+      "step": 5090
+    },
+    {
+      "epoch": 3.958090803259604,
+      "grad_norm": 0.7146425843238831,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 5100
+    },
+    {
+      "epoch": 3.965851765618937,
+      "grad_norm": 0.6555036902427673,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9736127279782694,
+      "grad_norm": 0.7037415504455566,
+      "learning_rate": 0.0002,
+      "loss": 1.4274,
+      "step": 5120
+    },
+    {
+      "epoch": 3.981373690337602,
+      "grad_norm": 0.7235575914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5130
+    },
+    {
+      "epoch": 3.9891346526969347,
+      "grad_norm": 0.7092325687408447,
+      "learning_rate": 0.0002,
+      "loss": 1.4455,
+      "step": 5140
+    },
+    {
+      "epoch": 3.996895615056267,
+      "grad_norm": 0.7490319609642029,
+      "learning_rate": 0.0002,
+      "loss": 1.4512,
+      "step": 5150
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9131355285644531,
+      "eval_runtime": 105.5778,
+      "eval_samples_per_second": 4.802,
+      "eval_steps_per_second": 0.606,
+      "step": 5154
+    },
+    {
+      "epoch": 4.0046565774155995,
+      "grad_norm": 0.7075854539871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 5160
+    },
+    {
+      "epoch": 4.012417539774932,
+      "grad_norm": 0.9466007351875305,
+      "learning_rate": 0.0002,
+      "loss": 1.209,
+      "step": 5170
+    },
+    {
+      "epoch": 4.020178502134264,
+      "grad_norm": 1.0297044515609741,
+      "learning_rate": 0.0002,
+      "loss": 1.2567,
+      "step": 5180
+    },
+    {
+      "epoch": 4.027939464493597,
+      "grad_norm": 0.7765059471130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5190
+    },
+    {
+      "epoch": 4.03570042685293,
+      "grad_norm": 0.995760977268219,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 5200
+    },
+    {
+      "epoch": 4.043461389212262,
+      "grad_norm": 0.8663829565048218,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 5210
+    },
+    {
+      "epoch": 4.051222351571595,
+      "grad_norm": 1.0660825967788696,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 5220
+    },
+    {
+      "epoch": 4.058983313930927,
+      "grad_norm": 0.9858174920082092,
+      "learning_rate": 0.0002,
+      "loss": 1.1676,
+      "step": 5230
+    },
+    {
+      "epoch": 4.06674427629026,
+      "grad_norm": 0.8911338448524475,
+      "learning_rate": 0.0002,
+      "loss": 1.2448,
+      "step": 5240
+    },
+    {
+      "epoch": 4.074505238649593,
+      "grad_norm": 1.0848394632339478,
+      "learning_rate": 0.0002,
+      "loss": 1.1858,
+      "step": 5250
+    },
+    {
+      "epoch": 4.082266201008925,
+      "grad_norm": 1.0849905014038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5260
+    },
+    {
+      "epoch": 4.090027163368258,
+      "grad_norm": 1.0497841835021973,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 5270
+    },
+    {
+      "epoch": 4.0977881257275905,
+      "grad_norm": 0.8943053483963013,
+      "learning_rate": 0.0002,
+      "loss": 1.2552,
+      "step": 5280
+    },
+    {
+      "epoch": 4.1055490880869225,
+      "grad_norm": 0.8432527184486389,
+      "learning_rate": 0.0002,
+      "loss": 1.1923,
+      "step": 5290
+    },
+    {
+      "epoch": 4.113310050446255,
+      "grad_norm": 0.9690414667129517,
+      "learning_rate": 0.0002,
+      "loss": 1.1634,
+      "step": 5300
+    },
+    {
+      "epoch": 4.121071012805588,
+      "grad_norm": 0.7790773510932922,
+      "learning_rate": 0.0002,
+      "loss": 1.3019,
+      "step": 5310
+    },
+    {
+      "epoch": 4.12883197516492,
+      "grad_norm": 0.9289211630821228,
+      "learning_rate": 0.0002,
+      "loss": 1.1806,
+      "step": 5320
+    },
+    {
+      "epoch": 4.136592937524253,
+      "grad_norm": 1.0785125494003296,
+      "learning_rate": 0.0002,
+      "loss": 1.1458,
+      "step": 5330
+    },
+    {
+      "epoch": 4.144353899883585,
+      "grad_norm": 0.8559591770172119,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5340
+    },
+    {
+      "epoch": 4.152114862242918,
+      "grad_norm": 0.9405956268310547,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5350
+    },
+    {
+      "epoch": 4.159875824602251,
+      "grad_norm": 0.9942827820777893,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 5360
+    },
+    {
+      "epoch": 4.167636786961583,
+      "grad_norm": 0.9141933917999268,
+      "learning_rate": 0.0002,
+      "loss": 1.1659,
+      "step": 5370
+    },
+    {
+      "epoch": 4.175397749320916,
+      "grad_norm": 0.8206015229225159,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 5380
+    },
+    {
+      "epoch": 4.183158711680249,
+      "grad_norm": 0.9340888857841492,
+      "learning_rate": 0.0002,
+      "loss": 1.2778,
+      "step": 5390
+    },
+    {
+      "epoch": 4.190919674039581,
+      "grad_norm": 1.2122114896774292,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 5400
+    },
+    {
+      "epoch": 4.1986806363989135,
+      "grad_norm": 1.0661298036575317,
+      "learning_rate": 0.0002,
+      "loss": 1.2371,
+      "step": 5410
+    },
+    {
+      "epoch": 4.206441598758246,
+      "grad_norm": 0.9372861385345459,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 5420
+    },
+    {
+      "epoch": 4.214202561117578,
+      "grad_norm": 0.894012987613678,
+      "learning_rate": 0.0002,
+      "loss": 1.2653,
+      "step": 5430
+    },
+    {
+      "epoch": 4.221963523476911,
+      "grad_norm": 1.0647753477096558,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5440
+    },
+    {
+      "epoch": 4.229724485836243,
+      "grad_norm": 0.989179790019989,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 5450
+    },
+    {
+      "epoch": 4.237485448195576,
+      "grad_norm": 1.1601181030273438,
+      "learning_rate": 0.0002,
+      "loss": 1.2715,
+      "step": 5460
+    },
+    {
+      "epoch": 4.245246410554909,
+      "grad_norm": 0.9395585656166077,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 5470
+    },
+    {
+      "epoch": 4.253007372914241,
+      "grad_norm": 0.9527766108512878,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 5480
+    },
+    {
+      "epoch": 4.260768335273574,
+      "grad_norm": 1.0319520235061646,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 5490
+    },
+    {
+      "epoch": 4.268529297632907,
+      "grad_norm": 0.8659824728965759,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 5500
+    },
+    {
+      "epoch": 4.276290259992239,
+      "grad_norm": 1.099211573600769,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 5510
+    },
+    {
+      "epoch": 4.284051222351572,
+      "grad_norm": 0.9363361597061157,
+      "learning_rate": 0.0002,
+      "loss": 1.2508,
+      "step": 5520
+    },
+    {
+      "epoch": 4.2918121847109045,
+      "grad_norm": 0.8437647223472595,
+      "learning_rate": 0.0002,
+      "loss": 1.189,
+      "step": 5530
+    },
+    {
+      "epoch": 4.2995731470702365,
+      "grad_norm": 0.9181258678436279,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 5540
+    },
+    {
+      "epoch": 4.307334109429569,
+      "grad_norm": 0.9059357643127441,
+      "learning_rate": 0.0002,
+      "loss": 1.2092,
+      "step": 5550
+    },
+    {
+      "epoch": 4.315095071788901,
+      "grad_norm": 0.9337241649627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2189,
+      "step": 5560
+    },
+    {
+      "epoch": 4.322856034148234,
+      "grad_norm": 0.9428889155387878,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 5570
+    },
+    {
+      "epoch": 4.330616996507567,
+      "grad_norm": 1.003589153289795,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 5580
+    },
+    {
+      "epoch": 4.338377958866899,
+      "grad_norm": 1.1249268054962158,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 5590
+    },
+    {
+      "epoch": 4.346138921226232,
+      "grad_norm": 0.8623469471931458,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 5600
+    },
+    {
+      "epoch": 4.353899883585565,
+      "grad_norm": 1.1389174461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5610
+    },
+    {
+      "epoch": 4.361660845944897,
+      "grad_norm": 1.0136264562606812,
+      "learning_rate": 0.0002,
+      "loss": 1.2245,
+      "step": 5620
+    },
+    {
+      "epoch": 4.36942180830423,
+      "grad_norm": 0.9567070603370667,
+      "learning_rate": 0.0002,
+      "loss": 1.3473,
+      "step": 5630
+    },
+    {
+      "epoch": 4.377182770663563,
+      "grad_norm": 1.0592148303985596,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 5640
+    },
+    {
+      "epoch": 4.384943733022895,
+      "grad_norm": 1.0110485553741455,
+      "learning_rate": 0.0002,
+      "loss": 1.212,
+      "step": 5650
+    },
+    {
+      "epoch": 4.3927046953822275,
+      "grad_norm": 0.9914907217025757,
+      "learning_rate": 0.0002,
+      "loss": 1.2086,
+      "step": 5660
+    },
+    {
+      "epoch": 4.4004656577415595,
+      "grad_norm": 0.9447247982025146,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 5670
+    },
+    {
+      "epoch": 4.408226620100892,
+      "grad_norm": 0.9644378423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 5680
+    },
+    {
+      "epoch": 4.415987582460225,
+      "grad_norm": 0.920676589012146,
+      "learning_rate": 0.0002,
+      "loss": 1.2773,
+      "step": 5690
+    },
+    {
+      "epoch": 4.423748544819557,
+      "grad_norm": 1.060570478439331,
+      "learning_rate": 0.0002,
+      "loss": 1.2792,
+      "step": 5700
+    },
+    {
+      "epoch": 4.43150950717889,
+      "grad_norm": 0.8857738971710205,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5710
+    },
+    {
+      "epoch": 4.439270469538223,
+      "grad_norm": 1.0536398887634277,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 5720
+    },
+    {
+      "epoch": 4.447031431897555,
+      "grad_norm": 0.990847110748291,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 5730
+    },
+    {
+      "epoch": 4.454792394256888,
+      "grad_norm": 0.9692499041557312,
+      "learning_rate": 0.0002,
+      "loss": 1.2469,
+      "step": 5740
+    },
+    {
+      "epoch": 4.462553356616221,
+      "grad_norm": 1.0376402139663696,
+      "learning_rate": 0.0002,
+      "loss": 1.2269,
+      "step": 5750
+    },
+    {
+      "epoch": 4.470314318975553,
+      "grad_norm": 1.3863259553909302,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 5760
+    },
+    {
+      "epoch": 4.478075281334886,
+      "grad_norm": 0.978379487991333,
+      "learning_rate": 0.0002,
+      "loss": 1.2591,
+      "step": 5770
+    },
+    {
+      "epoch": 4.485836243694218,
+      "grad_norm": 1.0973085165023804,
+      "learning_rate": 0.0002,
+      "loss": 1.2729,
+      "step": 5780
+    },
+    {
+      "epoch": 4.4935972060535505,
+      "grad_norm": 1.057006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 5790
+    },
+    {
+      "epoch": 4.501358168412883,
+      "grad_norm": 0.9247729182243347,
+      "learning_rate": 0.0002,
+      "loss": 1.2476,
+      "step": 5800
+    },
+    {
+      "epoch": 4.509119130772215,
+      "grad_norm": 1.0447787046432495,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 5810
+    },
+    {
+      "epoch": 4.516880093131548,
+      "grad_norm": 1.1930429935455322,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 5820
+    },
+    {
+      "epoch": 4.524641055490881,
+      "grad_norm": 0.9867590069770813,
+      "learning_rate": 0.0002,
+      "loss": 1.2596,
+      "step": 5830
+    },
+    {
+      "epoch": 4.532402017850213,
+      "grad_norm": 0.9591100215911865,
+      "learning_rate": 0.0002,
+      "loss": 1.2766,
+      "step": 5840
+    },
+    {
+      "epoch": 4.540162980209546,
+      "grad_norm": 0.9950753450393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 5850
+    },
+    {
+      "epoch": 4.547923942568879,
+      "grad_norm": 1.0087506771087646,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 5860
+    },
+    {
+      "epoch": 4.555684904928211,
+      "grad_norm": 1.0934417247772217,
+      "learning_rate": 0.0002,
+      "loss": 1.3165,
+      "step": 5870
+    },
+    {
+      "epoch": 4.563445867287544,
+      "grad_norm": 1.107987403869629,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 5880
+    },
+    {
+      "epoch": 4.571206829646876,
+      "grad_norm": 0.9147276878356934,
+      "learning_rate": 0.0002,
+      "loss": 1.2184,
+      "step": 5890
+    },
+    {
+      "epoch": 4.578967792006209,
+      "grad_norm": 1.036780595779419,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 5900
+    },
+    {
+      "epoch": 4.5867287543655415,
+      "grad_norm": 0.9284719824790955,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 5910
+    },
+    {
+      "epoch": 4.5944897167248735,
+      "grad_norm": 0.9141898155212402,
+      "learning_rate": 0.0002,
+      "loss": 1.3693,
+      "step": 5920
+    },
+    {
+      "epoch": 4.602250679084206,
+      "grad_norm": 1.0447357892990112,
+      "learning_rate": 0.0002,
+      "loss": 1.2319,
+      "step": 5930
+    },
+    {
+      "epoch": 4.610011641443539,
+      "grad_norm": 0.9309114217758179,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 5940
+    },
+    {
+      "epoch": 4.617772603802871,
+      "grad_norm": 1.2986129522323608,
+      "learning_rate": 0.0002,
+      "loss": 1.2827,
+      "step": 5950
+    },
+    {
+      "epoch": 4.625533566162204,
+      "grad_norm": 0.9221704602241516,
+      "learning_rate": 0.0002,
+      "loss": 1.312,
+      "step": 5960
+    },
+    {
+      "epoch": 4.633294528521537,
+      "grad_norm": 0.9228187799453735,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 5970
+    },
+    {
+      "epoch": 4.641055490880869,
+      "grad_norm": 0.9483116269111633,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 5980
+    },
+    {
+      "epoch": 4.648816453240202,
+      "grad_norm": 1.0218974351882935,
+      "learning_rate": 0.0002,
+      "loss": 1.3437,
+      "step": 5990
+    },
+    {
+      "epoch": 4.656577415599534,
+      "grad_norm": 0.9764600396156311,
+      "learning_rate": 0.0002,
+      "loss": 1.3085,
+      "step": 6000
+    },
+    {
+      "epoch": 4.664338377958867,
+      "grad_norm": 0.9115710258483887,
+      "learning_rate": 0.0002,
+      "loss": 1.197,
+      "step": 6010
+    },
+    {
+      "epoch": 4.6720993403182,
+      "grad_norm": 0.9245651364326477,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 6020
+    },
+    {
+      "epoch": 4.6798603026775325,
+      "grad_norm": 0.9686311483383179,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 6030
+    },
+    {
+      "epoch": 4.6876212650368645,
+      "grad_norm": 1.1807392835617065,
+      "learning_rate": 0.0002,
+      "loss": 1.2702,
+      "step": 6040
+    },
+    {
+      "epoch": 4.695382227396197,
+      "grad_norm": 1.0358641147613525,
+      "learning_rate": 0.0002,
+      "loss": 1.328,
+      "step": 6050
+    },
+    {
+      "epoch": 4.703143189755529,
+      "grad_norm": 0.987332284450531,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6060
+    },
+    {
+      "epoch": 4.710904152114862,
+      "grad_norm": 1.0526494979858398,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6070
+    },
+    {
+      "epoch": 4.718665114474195,
+      "grad_norm": 1.0276758670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.2246,
+      "step": 6080
+    },
+    {
+      "epoch": 4.726426076833527,
+      "grad_norm": 0.9904406666755676,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6090
+    },
+    {
+      "epoch": 4.73418703919286,
+      "grad_norm": 1.0084882974624634,
+      "learning_rate": 0.0002,
+      "loss": 1.2797,
+      "step": 6100
+    },
+    {
+      "epoch": 4.741948001552192,
+      "grad_norm": 0.8646450638771057,
+      "learning_rate": 0.0002,
+      "loss": 1.2656,
+      "step": 6110
+    },
+    {
+      "epoch": 4.749708963911525,
+      "grad_norm": 0.9233377575874329,
+      "learning_rate": 0.0002,
+      "loss": 1.3063,
+      "step": 6120
+    },
+    {
+      "epoch": 4.757469926270858,
+      "grad_norm": 0.9675140976905823,
+      "learning_rate": 0.0002,
+      "loss": 1.2642,
+      "step": 6130
+    },
+    {
+      "epoch": 4.765230888630191,
+      "grad_norm": 0.9639796018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6140
+    },
+    {
+      "epoch": 4.772991850989523,
+      "grad_norm": 0.925199568271637,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 6150
+    },
+    {
+      "epoch": 4.7807528133488555,
+      "grad_norm": 1.050901174545288,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 6160
+    },
+    {
+      "epoch": 4.7885137757081875,
+      "grad_norm": 0.8920623660087585,
+      "learning_rate": 0.0002,
+      "loss": 1.301,
+      "step": 6170
+    },
+    {
+      "epoch": 4.79627473806752,
+      "grad_norm": 0.8964757919311523,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6180
+    },
+    {
+      "epoch": 4.804035700426853,
+      "grad_norm": 1.0839070081710815,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 6190
+    },
+    {
+      "epoch": 4.811796662786185,
+      "grad_norm": 0.8809942007064819,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 6200
+    },
+    {
+      "epoch": 4.819557625145518,
+      "grad_norm": 1.0216195583343506,
+      "learning_rate": 0.0002,
+      "loss": 1.321,
+      "step": 6210
+    },
+    {
+      "epoch": 4.827318587504851,
+      "grad_norm": 0.892005980014801,
+      "learning_rate": 0.0002,
+      "loss": 1.3033,
+      "step": 6220
+    },
+    {
+      "epoch": 4.835079549864183,
+      "grad_norm": 0.9957166910171509,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6230
+    },
+    {
+      "epoch": 4.842840512223516,
+      "grad_norm": 0.9720533490180969,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 6240
+    },
+    {
+      "epoch": 4.850601474582849,
+      "grad_norm": 0.9336182475090027,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 6250
+    },
+    {
+      "epoch": 4.858362436942181,
+      "grad_norm": 1.2611457109451294,
+      "learning_rate": 0.0002,
+      "loss": 1.3136,
+      "step": 6260
+    },
+    {
+      "epoch": 4.866123399301514,
+      "grad_norm": 0.8927203416824341,
+      "learning_rate": 0.0002,
+      "loss": 1.2234,
+      "step": 6270
+    },
+    {
+      "epoch": 4.873884361660846,
+      "grad_norm": 0.9706710577011108,
+      "learning_rate": 0.0002,
+      "loss": 1.3463,
+      "step": 6280
+    },
+    {
+      "epoch": 4.8816453240201785,
+      "grad_norm": 1.1461690664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 6290
+    },
+    {
+      "epoch": 4.889406286379511,
+      "grad_norm": 0.9930381178855896,
+      "learning_rate": 0.0002,
+      "loss": 1.2566,
+      "step": 6300
+    },
+    {
+      "epoch": 4.897167248738843,
+      "grad_norm": 0.91451096534729,
+      "learning_rate": 0.0002,
+      "loss": 1.2568,
+      "step": 6310
+    },
+    {
+      "epoch": 4.904928211098176,
+      "grad_norm": 1.0319571495056152,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 6320
+    },
+    {
+      "epoch": 4.912689173457509,
+      "grad_norm": 0.990140438079834,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6330
+    },
+    {
+      "epoch": 4.920450135816841,
+      "grad_norm": 1.2466117143630981,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 6340
+    },
+    {
+      "epoch": 4.928211098176174,
+      "grad_norm": 1.0316979885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 6350
+    },
+    {
+      "epoch": 4.935972060535507,
+      "grad_norm": 1.0643759965896606,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 6360
+    },
+    {
+      "epoch": 4.943733022894839,
+      "grad_norm": 0.9703279733657837,
+      "learning_rate": 0.0002,
+      "loss": 1.2559,
+      "step": 6370
+    },
+    {
+      "epoch": 4.951493985254172,
+      "grad_norm": 0.9767927527427673,
+      "learning_rate": 0.0002,
+      "loss": 1.2155,
+      "step": 6380
+    },
+    {
+      "epoch": 4.959254947613504,
+      "grad_norm": 0.960854172706604,
+      "learning_rate": 0.0002,
+      "loss": 1.2437,
+      "step": 6390
+    },
+    {
+      "epoch": 4.967015909972837,
+      "grad_norm": 0.9922910332679749,
+      "learning_rate": 0.0002,
+      "loss": 1.3314,
+      "step": 6400
+    },
+    {
+      "epoch": 4.9747768723321695,
+      "grad_norm": 0.956470787525177,
+      "learning_rate": 0.0002,
+      "loss": 1.3018,
+      "step": 6410
+    },
+    {
+      "epoch": 4.9825378346915015,
+      "grad_norm": 0.9637242555618286,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 6420
+    },
+    {
+      "epoch": 4.990298797050834,
+      "grad_norm": 1.0855202674865723,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 6430
+    },
+    {
+      "epoch": 4.998059759410167,
+      "grad_norm": 0.9655316472053528,
+      "learning_rate": 0.0002,
+      "loss": 1.3015,
+      "step": 6440
+    },
+    {
+      "epoch": 4.9996119518820334,
+      "eval_loss": 2.0410802364349365,
+      "eval_runtime": 113.04,
+      "eval_samples_per_second": 4.485,
+      "eval_steps_per_second": 0.566,
+      "step": 6442
+    },
+    {
+      "epoch": 5.005820721769499,
+      "grad_norm": 1.1676199436187744,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 6450
+    },
+    {
+      "epoch": 5.013581684128832,
+      "grad_norm": 1.4317965507507324,
+      "learning_rate": 0.0002,
+      "loss": 1.041,
+      "step": 6460
+    },
+    {
+      "epoch": 5.021342646488165,
+      "grad_norm": 1.460443377494812,
+      "learning_rate": 0.0002,
+      "loss": 0.9546,
+      "step": 6470
+    },
+    {
+      "epoch": 5.029103608847497,
+      "grad_norm": 1.2299214601516724,
+      "learning_rate": 0.0002,
+      "loss": 1.0014,
+      "step": 6480
+    },
+    {
+      "epoch": 5.03686457120683,
+      "grad_norm": 1.3125724792480469,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 6490
+    },
+    {
+      "epoch": 5.044625533566162,
+      "grad_norm": 1.1252319812774658,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 6500
+    },
+    {
+      "epoch": 5.052386495925495,
+      "grad_norm": 0.9970866441726685,
+      "learning_rate": 0.0002,
+      "loss": 0.976,
+      "step": 6510
+    },
+    {
+      "epoch": 5.060147458284828,
+      "grad_norm": 1.229069709777832,
+      "learning_rate": 0.0002,
+      "loss": 0.9731,
+      "step": 6520
+    },
+    {
+      "epoch": 5.06790842064416,
+      "grad_norm": 1.2430938482284546,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 6530
+    },
+    {
+      "epoch": 5.0756693830034925,
+      "grad_norm": 1.0522737503051758,
+      "learning_rate": 0.0002,
+      "loss": 1.0236,
+      "step": 6540
+    },
+    {
+      "epoch": 5.083430345362825,
+      "grad_norm": 1.108890175819397,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 6550
+    },
+    {
+      "epoch": 5.091191307722157,
+      "grad_norm": 1.156912922859192,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 6560
+    },
+    {
+      "epoch": 5.09895227008149,
+      "grad_norm": 1.405895709991455,
+      "learning_rate": 0.0002,
+      "loss": 1.0415,
+      "step": 6570
+    },
+    {
+      "epoch": 5.106713232440823,
+      "grad_norm": 1.2005155086517334,
+      "learning_rate": 0.0002,
+      "loss": 0.9811,
+      "step": 6580
+    },
+    {
+      "epoch": 5.114474194800155,
+      "grad_norm": 1.181443452835083,
+      "learning_rate": 0.0002,
+      "loss": 0.9862,
+      "step": 6590
+    },
+    {
+      "epoch": 5.122235157159488,
+      "grad_norm": 2.3444771766662598,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 6600
+    },
+    {
+      "epoch": 5.12999611951882,
+      "grad_norm": 1.216988444328308,
+      "learning_rate": 0.0002,
+      "loss": 1.0455,
+      "step": 6610
+    },
+    {
+      "epoch": 5.137757081878153,
+      "grad_norm": 1.369553565979004,
+      "learning_rate": 0.0002,
+      "loss": 1.0549,
+      "step": 6620
+    },
+    {
+      "epoch": 5.145518044237486,
+      "grad_norm": 1.177964687347412,
+      "learning_rate": 0.0002,
+      "loss": 1.0056,
+      "step": 6630
+    },
+    {
+      "epoch": 5.153279006596818,
+      "grad_norm": 1.1397041082382202,
+      "learning_rate": 0.0002,
+      "loss": 1.1025,
+      "step": 6640
+    },
+    {
+      "epoch": 5.161039968956151,
+      "grad_norm": 1.3976861238479614,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 6650
+    },
+    {
+      "epoch": 5.1688009313154835,
+      "grad_norm": 1.4824495315551758,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 6660
+    },
+    {
+      "epoch": 5.1765618936748155,
+      "grad_norm": 1.2653018236160278,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 6670
+    },
+    {
+      "epoch": 5.184322856034148,
+      "grad_norm": 1.3106069564819336,
+      "learning_rate": 0.0002,
+      "loss": 0.9971,
+      "step": 6680
+    },
+    {
+      "epoch": 5.192083818393481,
+      "grad_norm": 1.3140279054641724,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 6690
+    },
+    {
+      "epoch": 5.199844780752813,
+      "grad_norm": 1.3900256156921387,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 6700
+    },
+    {
+      "epoch": 5.207605743112146,
+      "grad_norm": 1.3191124200820923,
+      "learning_rate": 0.0002,
+      "loss": 1.0285,
+      "step": 6710
+    },
+    {
+      "epoch": 5.215366705471478,
+      "grad_norm": 1.176107406616211,
+      "learning_rate": 0.0002,
+      "loss": 0.9921,
+      "step": 6720
+    },
+    {
+      "epoch": 5.223127667830811,
+      "grad_norm": 1.2364883422851562,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 6730
+    },
+    {
+      "epoch": 5.230888630190144,
+      "grad_norm": 1.343022108078003,
+      "learning_rate": 0.0002,
+      "loss": 0.9599,
+      "step": 6740
+    },
+    {
+      "epoch": 5.238649592549476,
+      "grad_norm": 1.2826898097991943,
+      "learning_rate": 0.0002,
+      "loss": 1.0342,
+      "step": 6750
+    },
+    {
+      "epoch": 5.246410554908809,
+      "grad_norm": 1.500257134437561,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 6760
+    },
+    {
+      "epoch": 5.254171517268142,
+      "grad_norm": 1.2605743408203125,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 6770
+    },
+    {
+      "epoch": 5.261932479627474,
+      "grad_norm": 1.2355525493621826,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6780
+    },
+    {
+      "epoch": 5.2696934419868064,
+      "grad_norm": 1.2845789194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.0436,
+      "step": 6790
+    },
+    {
+      "epoch": 5.277454404346139,
+      "grad_norm": 1.3696625232696533,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 6800
+    },
+    {
+      "epoch": 5.285215366705471,
+      "grad_norm": 1.4051260948181152,
+      "learning_rate": 0.0002,
+      "loss": 1.0991,
+      "step": 6810
+    },
+    {
+      "epoch": 5.292976329064804,
+      "grad_norm": 1.266725778579712,
+      "learning_rate": 0.0002,
+      "loss": 1.0987,
+      "step": 6820
+    },
+    {
+      "epoch": 5.300737291424136,
+      "grad_norm": 1.3475236892700195,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 6830
+    },
+    {
+      "epoch": 5.308498253783469,
+      "grad_norm": 1.54409921169281,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 6840
+    },
+    {
+      "epoch": 5.316259216142802,
+      "grad_norm": 1.2391985654830933,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 6850
+    },
+    {
+      "epoch": 5.324020178502134,
+      "grad_norm": 1.2435699701309204,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 6860
+    },
+    {
+      "epoch": 5.331781140861467,
+      "grad_norm": 1.8803037405014038,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6870
+    },
+    {
+      "epoch": 5.3395421032208,
+      "grad_norm": 1.4195542335510254,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 6880
+    },
+    {
+      "epoch": 5.347303065580132,
+      "grad_norm": 1.1853394508361816,
+      "learning_rate": 0.0002,
+      "loss": 1.0273,
+      "step": 6890
+    },
+    {
+      "epoch": 5.355064027939465,
+      "grad_norm": 1.4016530513763428,
+      "learning_rate": 0.0002,
+      "loss": 1.0668,
+      "step": 6900
+    },
+    {
+      "epoch": 5.3628249902987974,
+      "grad_norm": 1.294339895248413,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 6910
+    },
+    {
+      "epoch": 5.370585952658129,
+      "grad_norm": 1.2952708005905151,
+      "learning_rate": 0.0002,
+      "loss": 1.0724,
+      "step": 6920
+    },
+    {
+      "epoch": 5.378346915017462,
+      "grad_norm": 1.1361510753631592,
+      "learning_rate": 0.0002,
+      "loss": 1.0098,
+      "step": 6930
+    },
+    {
+      "epoch": 5.386107877376794,
+      "grad_norm": 1.125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.0796,
+      "step": 6940
+    },
+    {
+      "epoch": 5.393868839736127,
+      "grad_norm": 1.1453300714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 6950
+    },
+    {
+      "epoch": 5.40162980209546,
+      "grad_norm": 1.4542768001556396,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6960
+    },
+    {
+      "epoch": 5.409390764454792,
+      "grad_norm": 1.2360988855361938,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 6970
+    },
+    {
+      "epoch": 5.417151726814125,
+      "grad_norm": 1.2182754278182983,
+      "learning_rate": 0.0002,
+      "loss": 1.0631,
+      "step": 6980
+    },
+    {
+      "epoch": 5.424912689173458,
+      "grad_norm": 1.2018693685531616,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 6990
+    },
+    {
+      "epoch": 5.43267365153279,
+      "grad_norm": 1.346124291419983,
+      "learning_rate": 0.0002,
+      "loss": 1.108,
+      "step": 7000
+    },
+    {
+      "epoch": 5.440434613892123,
+      "grad_norm": 1.2534189224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.0534,
+      "step": 7010
+    },
+    {
+      "epoch": 5.448195576251456,
+      "grad_norm": 1.2033339738845825,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 7020
+    },
+    {
+      "epoch": 5.4559565386107876,
+      "grad_norm": 1.2788134813308716,
+      "learning_rate": 0.0002,
+      "loss": 1.0714,
+      "step": 7030
+    },
+    {
+      "epoch": 5.46371750097012,
+      "grad_norm": 1.2751542329788208,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 7040
+    },
+    {
+      "epoch": 5.471478463329452,
+      "grad_norm": 1.3237019777297974,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 7050
+    },
+    {
+      "epoch": 5.479239425688785,
+      "grad_norm": 1.4932852983474731,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 7060
+    },
+    {
+      "epoch": 5.487000388048118,
+      "grad_norm": 1.4003876447677612,
+      "learning_rate": 0.0002,
+      "loss": 1.0197,
+      "step": 7070
+    },
+    {
+      "epoch": 5.49476135040745,
+      "grad_norm": 1.404799461364746,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 7080
+    },
+    {
+      "epoch": 5.502522312766783,
+      "grad_norm": 1.4486982822418213,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 7090
+    },
+    {
+      "epoch": 5.510283275126116,
+      "grad_norm": 1.1713480949401855,
+      "learning_rate": 0.0002,
+      "loss": 1.0645,
+      "step": 7100
+    },
+    {
+      "epoch": 5.518044237485448,
+      "grad_norm": 1.4062601327896118,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 7110
+    },
+    {
+      "epoch": 5.525805199844781,
+      "grad_norm": 1.211629867553711,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 7120
+    },
+    {
+      "epoch": 5.533566162204114,
+      "grad_norm": 1.2523176670074463,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 7130
+    },
+    {
+      "epoch": 5.541327124563446,
+      "grad_norm": 1.4467198848724365,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 7140
+    },
+    {
+      "epoch": 5.5490880869227786,
+      "grad_norm": 1.5961614847183228,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 7150
+    },
+    {
+      "epoch": 5.5568490492821105,
+      "grad_norm": 1.320656418800354,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 7160
+    },
+    {
+      "epoch": 5.564610011641443,
+      "grad_norm": 1.2423332929611206,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 7170
+    },
+    {
+      "epoch": 5.572370974000776,
+      "grad_norm": 1.2919669151306152,
+      "learning_rate": 0.0002,
+      "loss": 1.0046,
+      "step": 7180
+    },
+    {
+      "epoch": 5.580131936360108,
+      "grad_norm": 1.1678385734558105,
+      "learning_rate": 0.0002,
+      "loss": 1.046,
+      "step": 7190
+    },
+    {
+      "epoch": 5.587892898719441,
+      "grad_norm": 1.4250764846801758,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 7200
+    },
+    {
+      "epoch": 5.595653861078774,
+      "grad_norm": 1.5308716297149658,
+      "learning_rate": 0.0002,
+      "loss": 1.1254,
+      "step": 7210
+    },
+    {
+      "epoch": 5.603414823438106,
+      "grad_norm": 1.2678815126419067,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7220
+    },
+    {
+      "epoch": 5.611175785797439,
+      "grad_norm": 1.127856969833374,
+      "learning_rate": 0.0002,
+      "loss": 1.0846,
+      "step": 7230
+    },
+    {
+      "epoch": 5.618936748156772,
+      "grad_norm": 1.3832560777664185,
+      "learning_rate": 0.0002,
+      "loss": 1.0647,
+      "step": 7240
+    },
+    {
+      "epoch": 5.626697710516104,
+      "grad_norm": 1.3226919174194336,
+      "learning_rate": 0.0002,
+      "loss": 1.0658,
+      "step": 7250
+    },
+    {
+      "epoch": 5.634458672875437,
+      "grad_norm": 1.3418006896972656,
+      "learning_rate": 0.0002,
+      "loss": 1.1175,
+      "step": 7260
+    },
+    {
+      "epoch": 5.642219635234769,
+      "grad_norm": 1.2625300884246826,
+      "learning_rate": 0.0002,
+      "loss": 1.0956,
+      "step": 7270
+    },
+    {
+      "epoch": 5.6499805975941015,
+      "grad_norm": 1.1579464673995972,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 7280
+    },
+    {
+      "epoch": 5.657741559953434,
+      "grad_norm": 1.4998650550842285,
+      "learning_rate": 0.0002,
+      "loss": 1.0447,
+      "step": 7290
+    },
+    {
+      "epoch": 5.665502522312766,
+      "grad_norm": 1.2670758962631226,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7300
+    },
+    {
+      "epoch": 5.673263484672099,
+      "grad_norm": 1.2959760427474976,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 7310
+    },
+    {
+      "epoch": 5.681024447031432,
+      "grad_norm": 1.2460671663284302,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 7320
+    },
+    {
+      "epoch": 5.688785409390764,
+      "grad_norm": 1.1313989162445068,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 7330
+    },
+    {
+      "epoch": 5.696546371750097,
+      "grad_norm": 1.282527208328247,
+      "learning_rate": 0.0002,
+      "loss": 1.0618,
+      "step": 7340
+    },
+    {
+      "epoch": 5.70430733410943,
+      "grad_norm": 1.3380206823349,
+      "learning_rate": 0.0002,
+      "loss": 1.1315,
+      "step": 7350
+    },
+    {
+      "epoch": 5.712068296468762,
+      "grad_norm": 1.1648279428482056,
+      "learning_rate": 0.0002,
+      "loss": 1.0949,
+      "step": 7360
+    },
+    {
+      "epoch": 5.719829258828095,
+      "grad_norm": 1.3059816360473633,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 7370
+    },
+    {
+      "epoch": 5.727590221187427,
+      "grad_norm": 1.1905046701431274,
+      "learning_rate": 0.0002,
+      "loss": 1.1496,
+      "step": 7380
+    },
+    {
+      "epoch": 5.73535118354676,
+      "grad_norm": 1.4089630842208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1356,
+      "step": 7390
+    },
+    {
+      "epoch": 5.7431121459060925,
+      "grad_norm": 1.256721019744873,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 7400
+    },
+    {
+      "epoch": 5.7508731082654245,
+      "grad_norm": 1.1915162801742554,
+      "learning_rate": 0.0002,
+      "loss": 1.0682,
+      "step": 7410
+    },
+    {
+      "epoch": 5.758634070624757,
+      "grad_norm": 1.1935480833053589,
+      "learning_rate": 0.0002,
+      "loss": 1.1257,
+      "step": 7420
+    },
+    {
+      "epoch": 5.76639503298409,
+      "grad_norm": 1.1761008501052856,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 7430
+    },
+    {
+      "epoch": 5.774155995343422,
+      "grad_norm": 1.2540549039840698,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 7440
+    },
+    {
+      "epoch": 5.781916957702755,
+      "grad_norm": 1.5295120477676392,
+      "learning_rate": 0.0002,
+      "loss": 1.1527,
+      "step": 7450
+    },
+    {
+      "epoch": 5.789677920062088,
+      "grad_norm": 1.1081160306930542,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 7460
+    },
+    {
+      "epoch": 5.79743888242142,
+      "grad_norm": 1.4381253719329834,
+      "learning_rate": 0.0002,
+      "loss": 1.1304,
+      "step": 7470
+    },
+    {
+      "epoch": 5.805199844780753,
+      "grad_norm": 1.3079341650009155,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 7480
+    },
+    {
+      "epoch": 5.812960807140085,
+      "grad_norm": 1.1372792720794678,
+      "learning_rate": 0.0002,
+      "loss": 1.0544,
+      "step": 7490
+    },
+    {
+      "epoch": 5.820721769499418,
+      "grad_norm": 1.3221744298934937,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 7500
+    },
+    {
+      "epoch": 5.828482731858751,
+      "grad_norm": 1.3436939716339111,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 7510
+    },
+    {
+      "epoch": 5.8362436942180835,
+      "grad_norm": 1.3916879892349243,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 7520
+    },
+    {
+      "epoch": 5.8440046565774155,
+      "grad_norm": 1.2463704347610474,
+      "learning_rate": 0.0002,
+      "loss": 1.0816,
+      "step": 7530
+    },
+    {
+      "epoch": 5.851765618936748,
+      "grad_norm": 1.097051739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.0745,
+      "step": 7540
+    },
+    {
+      "epoch": 5.85952658129608,
+      "grad_norm": 1.1554739475250244,
+      "learning_rate": 0.0002,
+      "loss": 1.1454,
+      "step": 7550
+    },
+    {
+      "epoch": 5.867287543655413,
+      "grad_norm": 1.2384694814682007,
+      "learning_rate": 0.0002,
+      "loss": 1.0953,
+      "step": 7560
+    },
+    {
+      "epoch": 5.875048506014746,
+      "grad_norm": 1.142815351486206,
+      "learning_rate": 0.0002,
+      "loss": 1.1734,
+      "step": 7570
+    },
+    {
+      "epoch": 5.882809468374078,
+      "grad_norm": 1.3637062311172485,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 7580
+    },
+    {
+      "epoch": 5.890570430733411,
+      "grad_norm": 1.2449073791503906,
+      "learning_rate": 0.0002,
+      "loss": 1.0781,
+      "step": 7590
+    },
+    {
+      "epoch": 5.898331393092743,
+      "grad_norm": 1.358058214187622,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 7600
+    },
+    {
+      "epoch": 5.906092355452076,
+      "grad_norm": 1.264655351638794,
+      "learning_rate": 0.0002,
+      "loss": 1.0779,
+      "step": 7610
+    },
+    {
+      "epoch": 5.913853317811409,
+      "grad_norm": 1.3186019659042358,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 7620
+    },
+    {
+      "epoch": 5.921614280170742,
+      "grad_norm": 1.4111460447311401,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 7630
+    },
+    {
+      "epoch": 5.929375242530074,
+      "grad_norm": 1.1078972816467285,
+      "learning_rate": 0.0002,
+      "loss": 1.1765,
+      "step": 7640
+    },
+    {
+      "epoch": 5.9371362048894065,
+      "grad_norm": 1.2742213010787964,
+      "learning_rate": 0.0002,
+      "loss": 1.1305,
+      "step": 7650
+    },
+    {
+      "epoch": 5.9448971672487385,
+      "grad_norm": 1.3412781953811646,
+      "learning_rate": 0.0002,
+      "loss": 1.144,
+      "step": 7660
+    },
+    {
+      "epoch": 5.952658129608071,
+      "grad_norm": 1.123005986213684,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 7670
+    },
+    {
+      "epoch": 5.960419091967404,
+      "grad_norm": 1.2203444242477417,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 7680
+    },
+    {
+      "epoch": 5.968180054326736,
+      "grad_norm": 1.341011643409729,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 7690
+    },
+    {
+      "epoch": 5.975941016686069,
+      "grad_norm": 1.2689454555511475,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 7700
+    },
+    {
+      "epoch": 5.983701979045401,
+      "grad_norm": 1.1518112421035767,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 7710
+    },
+    {
+      "epoch": 5.991462941404734,
+      "grad_norm": 1.3698320388793945,
+      "learning_rate": 0.0002,
+      "loss": 1.1868,
+      "step": 7720
+    },
+    {
+      "epoch": 5.999223903764067,
+      "grad_norm": 1.2812788486480713,
+      "learning_rate": 0.0002,
+      "loss": 1.0651,
+      "step": 7730
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.252762794494629,
+      "eval_runtime": 114.8471,
+      "eval_samples_per_second": 4.415,
+      "eval_steps_per_second": 0.557,
+      "step": 7731
+    },
+    {
+      "epoch": 6.006984866123399,
+      "grad_norm": 1.8642009496688843,
+      "learning_rate": 0.0002,
+      "loss": 0.8629,
+      "step": 7740
+    },
+    {
+      "epoch": 6.014745828482732,
+      "grad_norm": 1.7081232070922852,
+      "learning_rate": 0.0002,
+      "loss": 0.8435,
+      "step": 7750
+    },
+    {
+      "epoch": 6.022506790842065,
+      "grad_norm": 1.6233899593353271,
+      "learning_rate": 0.0002,
+      "loss": 0.7729,
+      "step": 7760
+    },
+    {
+      "epoch": 6.030267753201397,
+      "grad_norm": 1.5111888647079468,
+      "learning_rate": 0.0002,
+      "loss": 0.7907,
+      "step": 7770
+    },
+    {
+      "epoch": 6.0380287155607295,
+      "grad_norm": 1.5278418064117432,
+      "learning_rate": 0.0002,
+      "loss": 0.7908,
+      "step": 7780
+    },
+    {
+      "epoch": 6.045789677920062,
+      "grad_norm": 1.5932185649871826,
+      "learning_rate": 0.0002,
+      "loss": 0.835,
+      "step": 7790
+    },
+    {
+      "epoch": 6.053550640279394,
+      "grad_norm": 1.5990597009658813,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 7800
+    },
+    {
+      "epoch": 6.061311602638727,
+      "grad_norm": 1.7498669624328613,
+      "learning_rate": 0.0002,
+      "loss": 0.8559,
+      "step": 7810
+    },
+    {
+      "epoch": 6.06907256499806,
+      "grad_norm": 1.6105555295944214,
+      "learning_rate": 0.0002,
+      "loss": 0.8069,
+      "step": 7820
+    },
+    {
+      "epoch": 6.076833527357392,
+      "grad_norm": 1.5214293003082275,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 7830
+    },
+    {
+      "epoch": 6.084594489716725,
+      "grad_norm": 1.6586973667144775,
+      "learning_rate": 0.0002,
+      "loss": 0.8328,
+      "step": 7840
+    },
+    {
+      "epoch": 6.092355452076057,
+      "grad_norm": 1.467391848564148,
+      "learning_rate": 0.0002,
+      "loss": 0.8415,
+      "step": 7850
+    },
+    {
+      "epoch": 6.10011641443539,
+      "grad_norm": 1.537361741065979,
+      "learning_rate": 0.0002,
+      "loss": 0.8274,
+      "step": 7860
+    },
+    {
+      "epoch": 6.107877376794723,
+      "grad_norm": 1.621764898300171,
+      "learning_rate": 0.0002,
+      "loss": 0.8011,
+      "step": 7870
+    },
+    {
+      "epoch": 6.115638339154055,
+      "grad_norm": 1.583751916885376,
+      "learning_rate": 0.0002,
+      "loss": 0.8556,
+      "step": 7880
+    },
+    {
+      "epoch": 6.123399301513388,
+      "grad_norm": 1.6199619770050049,
+      "learning_rate": 0.0002,
+      "loss": 0.8829,
+      "step": 7890
+    },
+    {
+      "epoch": 6.1311602638727205,
+      "grad_norm": 1.6163095235824585,
+      "learning_rate": 0.0002,
+      "loss": 0.8226,
+      "step": 7900
+    },
+    {
+      "epoch": 6.1389212262320525,
+      "grad_norm": 1.6120976209640503,
+      "learning_rate": 0.0002,
+      "loss": 0.8203,
+      "step": 7910
+    },
+    {
+      "epoch": 6.146682188591385,
+      "grad_norm": 1.7886850833892822,
+      "learning_rate": 0.0002,
+      "loss": 0.7915,
+      "step": 7920
+    },
+    {
+      "epoch": 6.154443150950718,
+      "grad_norm": 1.408303141593933,
+      "learning_rate": 0.0002,
+      "loss": 0.7808,
+      "step": 7930
+    },
+    {
+      "epoch": 6.16220411331005,
+      "grad_norm": 1.6048113107681274,
+      "learning_rate": 0.0002,
+      "loss": 0.8404,
+      "step": 7940
+    },
+    {
+      "epoch": 6.169965075669383,
+      "grad_norm": 1.424306869506836,
+      "learning_rate": 0.0002,
+      "loss": 0.8705,
+      "step": 7950
+    },
+    {
+      "epoch": 6.177726038028716,
+      "grad_norm": 1.4453672170639038,
+      "learning_rate": 0.0002,
+      "loss": 0.8177,
+      "step": 7960
+    },
+    {
+      "epoch": 6.185487000388048,
+      "grad_norm": 1.3157061338424683,
+      "learning_rate": 0.0002,
+      "loss": 0.8182,
+      "step": 7970
+    },
+    {
+      "epoch": 6.193247962747381,
+      "grad_norm": 1.330541729927063,
+      "learning_rate": 0.0002,
+      "loss": 0.891,
+      "step": 7980
+    },
+    {
+      "epoch": 6.201008925106713,
+      "grad_norm": 1.6306229829788208,
+      "learning_rate": 0.0002,
+      "loss": 0.8599,
+      "step": 7990
+    },
+    {
+      "epoch": 6.208769887466046,
+      "grad_norm": 1.6332136392593384,
+      "learning_rate": 0.0002,
+      "loss": 0.9069,
+      "step": 8000
+    },
+    {
+      "epoch": 6.216530849825379,
+      "grad_norm": 1.708613395690918,
+      "learning_rate": 0.0002,
+      "loss": 0.83,
+      "step": 8010
+    },
+    {
+      "epoch": 6.224291812184711,
+      "grad_norm": 1.6637346744537354,
+      "learning_rate": 0.0002,
+      "loss": 0.8509,
+      "step": 8020
+    },
+    {
+      "epoch": 6.2320527745440435,
+      "grad_norm": 1.5675315856933594,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 8030
+    },
+    {
+      "epoch": 6.239813736903376,
+      "grad_norm": 1.5826327800750732,
+      "learning_rate": 0.0002,
+      "loss": 0.8491,
+      "step": 8040
+    },
+    {
+      "epoch": 6.247574699262708,
+      "grad_norm": 1.7382984161376953,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 8050
+    },
+    {
+      "epoch": 6.255335661622041,
+      "grad_norm": 1.5272295475006104,
+      "learning_rate": 0.0002,
+      "loss": 0.8795,
+      "step": 8060
+    },
+    {
+      "epoch": 6.263096623981374,
+      "grad_norm": 1.8195022344589233,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 8070
+    },
+    {
+      "epoch": 6.270857586340706,
+      "grad_norm": 1.679901361465454,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 8080
+    },
+    {
+      "epoch": 6.278618548700039,
+      "grad_norm": 1.4921348094940186,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 8090
+    },
+    {
+      "epoch": 6.286379511059371,
+      "grad_norm": 1.4627857208251953,
+      "learning_rate": 0.0002,
+      "loss": 0.899,
+      "step": 8100
+    },
+    {
+      "epoch": 6.294140473418704,
+      "grad_norm": 1.3528631925582886,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 8110
+    },
+    {
+      "epoch": 6.301901435778037,
+      "grad_norm": 1.6863102912902832,
+      "learning_rate": 0.0002,
+      "loss": 0.9355,
+      "step": 8120
+    },
+    {
+      "epoch": 6.309662398137369,
+      "grad_norm": 1.6178052425384521,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 8130
+    },
+    {
+      "epoch": 6.317423360496702,
+      "grad_norm": 1.7626280784606934,
+      "learning_rate": 0.0002,
+      "loss": 0.9182,
+      "step": 8140
+    },
+    {
+      "epoch": 6.3251843228560345,
+      "grad_norm": 1.7188845872879028,
+      "learning_rate": 0.0002,
+      "loss": 0.8886,
+      "step": 8150
+    },
+    {
+      "epoch": 6.3329452852153665,
+      "grad_norm": 1.5777133703231812,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 8160
+    },
+    {
+      "epoch": 6.340706247574699,
+      "grad_norm": 1.7653207778930664,
+      "learning_rate": 0.0002,
+      "loss": 0.9247,
+      "step": 8170
+    },
+    {
+      "epoch": 6.348467209934032,
+      "grad_norm": 1.6861237287521362,
+      "learning_rate": 0.0002,
+      "loss": 0.8003,
+      "step": 8180
+    },
+    {
+      "epoch": 6.356228172293364,
+      "grad_norm": 1.6318124532699585,
+      "learning_rate": 0.0002,
+      "loss": 0.884,
+      "step": 8190
+    },
+    {
+      "epoch": 6.363989134652697,
+      "grad_norm": 1.6192939281463623,
+      "learning_rate": 0.0002,
+      "loss": 0.8341,
+      "step": 8200
+    },
+    {
+      "epoch": 6.371750097012029,
+      "grad_norm": 1.7641773223876953,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 8210
+    },
+    {
+      "epoch": 6.379511059371362,
+      "grad_norm": 1.6470493078231812,
+      "learning_rate": 0.0002,
+      "loss": 0.8582,
+      "step": 8220
+    },
+    {
+      "epoch": 6.387272021730695,
+      "grad_norm": 1.5898468494415283,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 8230
+    },
+    {
+      "epoch": 6.395032984090027,
+      "grad_norm": 1.8025981187820435,
+      "learning_rate": 0.0002,
+      "loss": 0.9658,
+      "step": 8240
+    },
+    {
+      "epoch": 6.40279394644936,
+      "grad_norm": 1.7035106420516968,
+      "learning_rate": 0.0002,
+      "loss": 0.8953,
+      "step": 8250
+    },
+    {
+      "epoch": 6.410554908808693,
+      "grad_norm": 1.5968799591064453,
+      "learning_rate": 0.0002,
+      "loss": 0.9193,
+      "step": 8260
+    },
+    {
+      "epoch": 6.418315871168025,
+      "grad_norm": 1.7492800951004028,
+      "learning_rate": 0.0002,
+      "loss": 0.929,
+      "step": 8270
+    },
+    {
+      "epoch": 6.4260768335273575,
+      "grad_norm": 1.6914138793945312,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 8280
+    },
+    {
+      "epoch": 6.43383779588669,
+      "grad_norm": 1.5761380195617676,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 8290
+    },
+    {
+      "epoch": 6.441598758246022,
+      "grad_norm": 1.5164411067962646,
+      "learning_rate": 0.0002,
+      "loss": 0.8761,
+      "step": 8300
+    },
+    {
+      "epoch": 6.449359720605355,
+      "grad_norm": 1.6600215435028076,
+      "learning_rate": 0.0002,
+      "loss": 0.88,
+      "step": 8310
+    },
+    {
+      "epoch": 6.457120682964687,
+      "grad_norm": 1.2477679252624512,
+      "learning_rate": 0.0002,
+      "loss": 0.9113,
+      "step": 8320
+    },
+    {
+      "epoch": 6.46488164532402,
+      "grad_norm": 1.3698599338531494,
+      "learning_rate": 0.0002,
+      "loss": 0.8822,
+      "step": 8330
+    },
+    {
+      "epoch": 6.472642607683353,
+      "grad_norm": 1.4847341775894165,
+      "learning_rate": 0.0002,
+      "loss": 0.9295,
+      "step": 8340
+    },
+    {
+      "epoch": 6.480403570042685,
+      "grad_norm": 1.4713412523269653,
+      "learning_rate": 0.0002,
+      "loss": 0.9243,
+      "step": 8350
+    },
+    {
+      "epoch": 6.488164532402018,
+      "grad_norm": 1.334523320198059,
+      "learning_rate": 0.0002,
+      "loss": 0.9102,
+      "step": 8360
+    },
+    {
+      "epoch": 6.495925494761351,
+      "grad_norm": 2.0054359436035156,
+      "learning_rate": 0.0002,
+      "loss": 0.8563,
+      "step": 8370
+    },
+    {
+      "epoch": 6.503686457120683,
+      "grad_norm": 1.560014247894287,
+      "learning_rate": 0.0002,
+      "loss": 0.9759,
+      "step": 8380
+    },
+    {
+      "epoch": 6.511447419480016,
+      "grad_norm": 1.518526554107666,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 8390
+    },
+    {
+      "epoch": 6.5192083818393485,
+      "grad_norm": 1.3841272592544556,
+      "learning_rate": 0.0002,
+      "loss": 0.937,
+      "step": 8400
+    },
+    {
+      "epoch": 6.5269693441986805,
+      "grad_norm": 1.5191527605056763,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 8410
+    },
+    {
+      "epoch": 6.534730306558013,
+      "grad_norm": 1.5275579690933228,
+      "learning_rate": 0.0002,
+      "loss": 0.8899,
+      "step": 8420
+    },
+    {
+      "epoch": 6.542491268917345,
+      "grad_norm": 1.621590256690979,
+      "learning_rate": 0.0002,
+      "loss": 0.9291,
+      "step": 8430
+    },
+    {
+      "epoch": 6.550252231276678,
+      "grad_norm": 1.7939082384109497,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 8440
+    },
+    {
+      "epoch": 6.558013193636011,
+      "grad_norm": 1.4542964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8896,
+      "step": 8450
+    },
+    {
+      "epoch": 6.565774155995343,
+      "grad_norm": 1.5458455085754395,
+      "learning_rate": 0.0002,
+      "loss": 0.9393,
+      "step": 8460
+    },
+    {
+      "epoch": 6.573535118354676,
+      "grad_norm": 1.550359845161438,
+      "learning_rate": 0.0002,
+      "loss": 0.9028,
+      "step": 8470
+    },
+    {
+      "epoch": 6.581296080714009,
+      "grad_norm": 1.527757167816162,
+      "learning_rate": 0.0002,
+      "loss": 0.9271,
+      "step": 8480
+    },
+    {
+      "epoch": 6.589057043073341,
+      "grad_norm": 1.4683486223220825,
+      "learning_rate": 0.0002,
+      "loss": 0.966,
+      "step": 8490
+    },
+    {
+      "epoch": 6.596818005432674,
+      "grad_norm": 1.5057084560394287,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 8500
+    },
+    {
+      "epoch": 6.604578967792007,
+      "grad_norm": 1.648289442062378,
+      "learning_rate": 0.0002,
+      "loss": 0.9235,
+      "step": 8510
+    },
+    {
+      "epoch": 6.612339930151339,
+      "grad_norm": 1.578914761543274,
+      "learning_rate": 0.0002,
+      "loss": 0.9113,
+      "step": 8520
+    },
+    {
+      "epoch": 6.6201008925106715,
+      "grad_norm": 1.5064080953598022,
+      "learning_rate": 0.0002,
+      "loss": 0.8894,
+      "step": 8530
+    },
+    {
+      "epoch": 6.6278618548700035,
+      "grad_norm": 1.5717744827270508,
+      "learning_rate": 0.0002,
+      "loss": 0.8981,
+      "step": 8540
+    },
+    {
+      "epoch": 6.635622817229336,
+      "grad_norm": 1.7954767942428589,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 8550
+    },
+    {
+      "epoch": 6.643383779588669,
+      "grad_norm": 1.6172343492507935,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 8560
+    },
+    {
+      "epoch": 6.651144741948001,
+      "grad_norm": 1.6627886295318604,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 8570
+    },
+    {
+      "epoch": 6.658905704307334,
+      "grad_norm": 1.5264919996261597,
+      "learning_rate": 0.0002,
+      "loss": 0.959,
+      "step": 8580
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.609248161315918,
+      "learning_rate": 0.0002,
+      "loss": 0.9103,
+      "step": 8590
+    },
+    {
+      "epoch": 6.674427629025999,
+      "grad_norm": 1.5474581718444824,
+      "learning_rate": 0.0002,
+      "loss": 0.9395,
+      "step": 8600
+    },
+    {
+      "epoch": 6.682188591385332,
+      "grad_norm": 1.6294898986816406,
+      "learning_rate": 0.0002,
+      "loss": 0.9018,
+      "step": 8610
+    },
+    {
+      "epoch": 6.689949553744665,
+      "grad_norm": 1.612615942955017,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 8620
+    },
+    {
+      "epoch": 6.697710516103997,
+      "grad_norm": 1.741325855255127,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 8630
+    },
+    {
+      "epoch": 6.70547147846333,
+      "grad_norm": 1.5089004039764404,
+      "learning_rate": 0.0002,
+      "loss": 1.0475,
+      "step": 8640
+    },
+    {
+      "epoch": 6.713232440822662,
+      "grad_norm": 1.4725582599639893,
+      "learning_rate": 0.0002,
+      "loss": 1.0009,
+      "step": 8650
+    },
+    {
+      "epoch": 6.7209934031819945,
+      "grad_norm": 1.6992095708847046,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 8660
+    },
+    {
+      "epoch": 6.728754365541327,
+      "grad_norm": 1.5938470363616943,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 8670
+    },
+    {
+      "epoch": 6.736515327900659,
+      "grad_norm": 1.58723783493042,
+      "learning_rate": 0.0002,
+      "loss": 0.9411,
+      "step": 8680
+    },
+    {
+      "epoch": 6.744276290259992,
+      "grad_norm": 1.514389991760254,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 8690
+    },
+    {
+      "epoch": 6.752037252619325,
+      "grad_norm": 1.6799157857894897,
+      "learning_rate": 0.0002,
+      "loss": 0.9283,
+      "step": 8700
+    },
+    {
+      "epoch": 6.759798214978657,
+      "grad_norm": 1.5436359643936157,
+      "learning_rate": 0.0002,
+      "loss": 0.9138,
+      "step": 8710
+    },
+    {
+      "epoch": 6.76755917733799,
+      "grad_norm": 1.477137565612793,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8720
+    },
+    {
+      "epoch": 6.775320139697323,
+      "grad_norm": 1.7383503913879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0044,
+      "step": 8730
+    },
+    {
+      "epoch": 6.783081102056655,
+      "grad_norm": 1.8000324964523315,
+      "learning_rate": 0.0002,
+      "loss": 0.9492,
+      "step": 8740
+    },
+    {
+      "epoch": 6.790842064415988,
+      "grad_norm": 1.3099453449249268,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 8750
+    },
+    {
+      "epoch": 6.79860302677532,
+      "grad_norm": 1.8775172233581543,
+      "learning_rate": 0.0002,
+      "loss": 0.9709,
+      "step": 8760
+    },
+    {
+      "epoch": 6.806363989134653,
+      "grad_norm": 1.5832085609436035,
+      "learning_rate": 0.0002,
+      "loss": 0.9356,
+      "step": 8770
+    },
+    {
+      "epoch": 6.8141249514939854,
+      "grad_norm": 1.4903252124786377,
+      "learning_rate": 0.0002,
+      "loss": 0.9397,
+      "step": 8780
+    },
+    {
+      "epoch": 6.821885913853317,
+      "grad_norm": 1.6360470056533813,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 8790
+    },
+    {
+      "epoch": 6.82964687621265,
+      "grad_norm": 1.5457707643508911,
+      "learning_rate": 0.0002,
+      "loss": 0.957,
+      "step": 8800
+    },
+    {
+      "epoch": 6.837407838571983,
+      "grad_norm": 1.5449066162109375,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 8810
+    },
+    {
+      "epoch": 6.845168800931315,
+      "grad_norm": 1.618337631225586,
+      "learning_rate": 0.0002,
+      "loss": 1.0007,
+      "step": 8820
+    },
+    {
+      "epoch": 6.852929763290648,
+      "grad_norm": 1.38296639919281,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 8830
+    },
+    {
+      "epoch": 6.860690725649981,
+      "grad_norm": 1.6427991390228271,
+      "learning_rate": 0.0002,
+      "loss": 0.9349,
+      "step": 8840
+    },
+    {
+      "epoch": 6.868451688009313,
+      "grad_norm": 1.4980270862579346,
+      "learning_rate": 0.0002,
+      "loss": 1.0194,
+      "step": 8850
+    },
+    {
+      "epoch": 6.876212650368646,
+      "grad_norm": 1.3800020217895508,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 8860
+    },
+    {
+      "epoch": 6.883973612727978,
+      "grad_norm": 1.5971838235855103,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 8870
+    },
+    {
+      "epoch": 6.891734575087311,
+      "grad_norm": 1.4429489374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.0105,
+      "step": 8880
+    },
+    {
+      "epoch": 6.899495537446644,
+      "grad_norm": 1.4959166049957275,
+      "learning_rate": 0.0002,
+      "loss": 0.9143,
+      "step": 8890
+    },
+    {
+      "epoch": 6.907256499805976,
+      "grad_norm": 1.5776222944259644,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 8900
+    },
+    {
+      "epoch": 6.915017462165308,
+      "grad_norm": 1.510412573814392,
+      "learning_rate": 0.0002,
+      "loss": 0.9256,
+      "step": 8910
+    },
+    {
+      "epoch": 6.922778424524641,
+      "grad_norm": 1.7216295003890991,
+      "learning_rate": 0.0002,
+      "loss": 1.0095,
+      "step": 8920
+    },
+    {
+      "epoch": 6.930539386883973,
+      "grad_norm": 1.830762505531311,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 8930
+    },
+    {
+      "epoch": 6.938300349243306,
+      "grad_norm": 1.3472434282302856,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 8940
+    },
+    {
+      "epoch": 6.946061311602639,
+      "grad_norm": 1.5748040676116943,
+      "learning_rate": 0.0002,
+      "loss": 0.9718,
+      "step": 8950
+    },
+    {
+      "epoch": 6.953822273961971,
+      "grad_norm": 1.5317506790161133,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 8960
+    },
+    {
+      "epoch": 6.961583236321304,
+      "grad_norm": 1.5565721988677979,
+      "learning_rate": 0.0002,
+      "loss": 0.9513,
+      "step": 8970
+    },
+    {
+      "epoch": 6.969344198680636,
+      "grad_norm": 1.5288970470428467,
+      "learning_rate": 0.0002,
+      "loss": 0.9118,
+      "step": 8980
+    },
+    {
+      "epoch": 6.977105161039969,
+      "grad_norm": 1.562624454498291,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 8990
+    },
+    {
+      "epoch": 6.984866123399302,
+      "grad_norm": 1.3777633905410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9929,
+      "step": 9000
+    },
+    {
+      "epoch": 6.992627085758635,
+      "grad_norm": 1.5868972539901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 9010
+    },
+    {
+      "epoch": 6.9996119518820334,
+      "eval_loss": 2.4372169971466064,
+      "eval_runtime": 113.8966,
+      "eval_samples_per_second": 4.451,
+      "eval_steps_per_second": 0.562,
+      "step": 9019
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10304,
+  "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": 4.632753036931891e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-9019/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4a305d1de4d8f47c0252b4d7fe65a10dd8e2c22
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8ca6e2c3ac58fa2af9f99747566f932f41a5a4d5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7feb06ff53d5bf79374054a25b662309e705a2ca08dfa3b0bce7b8b4632fae
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_log.jsonl b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..14a0a3d0bef9090dc1477c17625532da9ecc79a5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/training_log.jsonl
@@ -0,0 +1,10 @@
+{"epoch": 0.9996119518820333, "step": 1288, "epoch_duration": 4887.0092253685, "total_accumulated_duration": 4887.0092253685, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12786.0}, "peak_memory_reserved": {"GPU_0": 12786.0}, "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}]}
+{"epoch": 2.0, "step": 2577, "epoch_duration": 4024.7662012577057, "total_accumulated_duration": 8911.775426626205, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 114.9794, "eval_samples_per_second": 4.409, "eval_steps_per_second": 0.557, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}]}
+{"epoch": 2.9996119518820334, "step": 3865, "epoch_duration": 2179.5441098213196, "total_accumulated_duration": 11091.319536447525, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520{"epoch": 0.9996119518820333, "step": 1288, "epoch_duration": 3340.132976293564, "total_accumulated_duration": 3340.132976293564, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12786.0}, "peak_memory_reserved": {"GPU_0": 12786.0}, "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}]}
+{"epoch": 2.0, "step": 2577, "epoch_duration": 3629.9410111904144, "total_accumulated_duration": 6970.073987483978, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-1288", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}]}
+{"epoch": 2.9996119518820334, "step": 3865, "epoch_duration": 3652.9453053474426, "total_accumulated_duration": 10623.01929283142, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}]}
+{"epoch": 4.0, "step": 5154, "epoch_duration": 3651.1835963726044, "total_accumulated_duration": 14274.202889204025, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}, {"eval_loss": 1.8472607135772705, "eval_runtime": 106.5541, "eval_samples_per_second": 4.758, "eval_steps_per_second": 0.601, "epoch": 2.9996119518820334, "step": 3865}, {"loss": 1.4983, "grad_norm": 0.5645653605461121, "learning_rate": 0.0002, "epoch": 3.0034924330616994, "step": 3870}, {"loss": 1.4334, "grad_norm": 0.6457151174545288, "learning_rate": 0.0002, "epoch": 3.0112533954210323, "step": 3880}, {"loss": 1.3899, "grad_norm": 0.583838164806366, "learning_rate": 0.0002, "epoch": 3.0190143577803648, "step": 3890}, {"loss": 1.3258, "grad_norm": 0.6819260120391846, "learning_rate": 0.0002, "epoch": 3.026775320139697, "step": 3900}, {"loss": 1.3458, "grad_norm": 0.6692903637886047, "learning_rate": 0.0002, "epoch": 3.03453628249903, "step": 3910}, {"loss": 1.4356, "grad_norm": 0.6101024746894836, "learning_rate": 0.0002, "epoch": 3.0422972448583625, "step": 3920}, {"loss": 1.394, "grad_norm": 0.7014093399047852, "learning_rate": 0.0002, "epoch": 3.050058207217695, "step": 3930}, {"loss": 1.3885, "grad_norm": 0.7380381226539612, "learning_rate": 0.0002, "epoch": 3.0578191695770274, "step": 3940}, {"loss": 1.4206, "grad_norm": 0.6607900857925415, "learning_rate": 0.0002, "epoch": 3.0655801319363603, "step": 3950}, {"loss": 1.4293, "grad_norm": 0.735263466835022, "learning_rate": 0.0002, "epoch": 3.0733410942956927, "step": 3960}, {"loss": 1.3966, "grad_norm": 0.6788513660430908, "learning_rate": 0.0002, "epoch": 3.081102056655025, "step": 3970}, {"loss": 1.3435, "grad_norm": 0.6347652673721313, "learning_rate": 0.0002, "epoch": 3.088863019014358, "step": 3980}, {"loss": 1.4518, "grad_norm": 0.7056642770767212, "learning_rate": 0.0002, "epoch": 3.0966239813736904, "step": 3990}, {"loss": 1.4474, "grad_norm": 0.6387075185775757, "learning_rate": 0.0002, "epoch": 3.104384943733023, "step": 4000}, {"loss": 1.3833, "grad_norm": 0.6701116561889648, "learning_rate": 0.0002, "epoch": 3.1121459060923553, "step": 4010}, {"loss": 1.404, "grad_norm": 0.7558449506759644, "learning_rate": 0.0002, "epoch": 3.119906868451688, "step": 4020}, {"loss": 1.3294, "grad_norm": 0.6612881422042847, "learning_rate": 0.0002, "epoch": 3.1276678308110206, "step": 4030}, {"loss": 1.439, "grad_norm": 0.7474587559700012, "learning_rate": 0.0002, "epoch": 3.135428793170353, "step": 4040}, {"loss": 1.4616, "grad_norm": 0.7292373776435852, "learning_rate": 0.0002, "epoch": 3.1431897555296855, "step": 4050}, {"loss": 1.3908, "grad_norm": 0.7432886958122253, "learning_rate": 0.0002, "epoch": 3.1509507178890184, "step": 4060}, {"loss": 1.4214, "grad_norm": 0.6366098523139954, "learning_rate": 0.0002, "epoch": 3.158711680248351, "step": 4070}, {"loss": 1.5044, "grad_norm": 0.6837611794471741, "learning_rate": 0.0002, "epoch": 3.1664726426076832, "step": 4080}, {"loss": 1.4332, "grad_norm": 0.7194393277168274, "learning_rate": 0.0002, "epoch": 3.174233604967016, "step": 4090}, {"loss": 1.3628, "grad_norm": 0.6963607668876648, "learning_rate": 0.0002, "epoch": 3.1819945673263486, "step": 4100}, {"loss": 1.4127, "grad_norm": 0.6404902935028076, "learning_rate": 0.0002, "epoch": 3.189755529685681, "step": 4110}, {"loss": 1.4394, "grad_norm": 0.7172070741653442, "learning_rate": 0.0002, "epoch": 3.1975164920450134, "step": 4120}, {"loss": 1.4658, "grad_norm": 0.6577759385108948, "learning_rate": 0.0002, "epoch": 3.2052774544043463, "step": 4130}, {"loss": 1.4019, "grad_norm": 0.6658480167388916, "learning_rate": 0.0002, "epoch": 3.2130384167636787, "step": 4140}, {"loss": 1.4348, "grad_norm": 0.6771699786186218, "learning_rate": 0.0002, "epoch": 3.220799379123011, "step": 4150}, {"loss": 1.4736, "grad_norm": 0.699035108089447, "learning_rate": 0.0002, "epoch": 3.2285603414823436, "step": 4160}, {"loss": 1.4096, "grad_norm": 0.7218514680862427, "learning_rate": 0.0002, "epoch": 3.2363213038416765, "step": 4170}, {"loss": 1.3637, "grad_norm": 0.6270631551742554, "learning_rate": 0.0002, "epoch": 3.244082266201009, "step": 4180}, {"loss": 1.4076, "grad_norm": 0.6828921437263489, "learning_rate": 0.0002, "epoch": 3.2518432285603414, "step": 4190}, {"loss": 1.4663, "grad_norm": 0.6005498170852661, "learning_rate": 0.0002, "epoch": 3.2596041909196742, "step": 4200}, {"loss": 1.4798, "grad_norm": 0.6974790692329407, "learning_rate": 0.0002, "epoch": 3.2673651532790067, "step": 4210}, {"loss": 1.5012, "grad_norm": 0.7269543409347534, "learning_rate": 0.0002, "epoch": 3.275126115638339, "step": 4220}, {"loss": 1.3848, "grad_norm": 0.6728787422180176, "learning_rate": 0.0002, "epoch": 3.2828870779976715, "step": 4230}, {"loss": 1.4112, "grad_norm": 0.676972508430481, "learning_rate": 0.0002, "epoch": 3.2906480403570044, "step": 4240}, {"loss": 1.4206, "grad_norm": 0.748309314250946, "learning_rate": 0.0002, "epoch": 3.298409002716337, "step": 4250}, {"loss": 1.4973, "grad_norm": 0.6976589560508728, "learning_rate": 0.0002, "epoch": 3.3061699650756693, "step": 4260}, {"loss": 1.3967, "grad_norm": 0.649780809879303, "learning_rate": 0.0002, "epoch": 3.3139309274350017, "step": 4270}, {"loss": 1.327, "grad_norm": 0.6529902815818787, "learning_rate": 0.0002, "epoch": 3.3216918897943346, "step": 4280}, {"loss": 1.4888, "grad_norm": 0.9273163676261902, "learning_rate": 0.0002, "epoch": 3.329452852153667, "step": 4290}, {"loss": 1.4859, "grad_norm": 0.717024028301239, "learning_rate": 0.0002, "epoch": 3.3372138145129995, "step": 4300}, {"loss": 1.4441, "grad_norm": 0.7914950251579285, "learning_rate": 0.0002, "epoch": 3.3449747768723324, "step": 4310}, {"loss": 1.432, "grad_norm": 0.7133203148841858, "learning_rate": 0.0002, "epoch": 3.352735739231665, "step": 4320}, {"loss": 1.4662, "grad_norm": 0.7409568428993225, "learning_rate": 0.0002, "epoch": 3.3604967015909972, "step": 4330}, {"loss": 1.3992, "grad_norm": 0.6993981003761292, "learning_rate": 0.0002, "epoch": 3.3682576639503297, "step": 4340}, {"loss": 1.4261, "grad_norm": 0.7114535570144653, "learning_rate": 0.0002, "epoch": 3.3760186263096625, "step": 4350}, {"loss": 1.4227, "grad_norm": 0.6790860295295715, "learning_rate": 0.0002, "epoch": 3.383779588668995, "step": 4360}, {"loss": 1.4128, "grad_norm": 0.6507849097251892, "learning_rate": 0.0002, "epoch": 3.3915405510283274, "step": 4370}, {"loss": 1.4559, "grad_norm": 0.5967804193496704, "learning_rate": 0.0002, "epoch": 3.39930151338766, "step": 4380}, {"loss": 1.3687, "grad_norm": 0.6625847816467285, "learning_rate": 0.0002, "epoch": 3.4070624757469927, "step": 4390}, {"loss": 1.4193, "grad_norm": 0.6736508011817932, "learning_rate": 0.0002, "epoch": 3.414823438106325, "step": 4400}, {"loss": 1.4363, "grad_norm": 0.7870860695838928, "learning_rate": 0.0002, "epoch": 3.4225844004656576, "step": 4410}, {"loss": 1.4114, "grad_norm": 0.7205295562744141, "learning_rate": 0.0002, "epoch": 3.4303453628249905, "step": 4420}, {"loss": 1.4131, "grad_norm": 0.6634634137153625, "learning_rate": 0.0002, "epoch": 3.438106325184323, "step": 4430}, {"loss": 1.4683, "grad_norm": 0.7562733292579651, "learning_rate": 0.0002, "epoch": 3.4458672875436553, "step": 4440}, {"loss": 1.3486, "grad_norm": 0.6585879921913147, "learning_rate": 0.0002, "epoch": 3.453628249902988, "step": 4450}, {"loss": 1.4283, "grad_norm": 0.6896792054176331, "learning_rate": 0.0002, "epoch": 3.4613892122623207, "step": 4460}, {"loss": 1.4208, "grad_norm": 0.6520342230796814, "learning_rate": 0.0002, "epoch": 3.469150174621653, "step": 4470}, {"loss": 1.3423, "grad_norm": 0.6760806441307068, "learning_rate": 0.0002, "epoch": 3.4769111369809855, "step": 4480}, {"loss": 1.4398, "grad_norm": 0.7539774179458618, "learning_rate": 0.0002, "epoch": 3.484672099340318, "step": 4490}, {"loss": 1.4534, "grad_norm": 0.7409411668777466, "learning_rate": 0.0002, "epoch": 3.492433061699651, "step": 4500}, {"loss": 1.4069, "grad_norm": 0.6876253485679626, "learning_rate": 0.0002, "epoch": 3.5001940240589833, "step": 4510}, {"loss": 1.4228, "grad_norm": 0.7028461694717407, "learning_rate": 0.0002, "epoch": 3.5079549864183157, "step": 4520}, {"loss": 1.4723, "grad_norm": 0.8056529760360718, "learning_rate": 0.0002, "epoch": 3.5157159487776486, "step": 4530}, {"loss": 1.4148, "grad_norm": 0.711338996887207, "learning_rate": 0.0002, "epoch": 3.523476911136981, "step": 4540}, {"loss": 1.5247, "grad_norm": 0.7343552708625793, "learning_rate": 0.0002, "epoch": 3.5312378734963135, "step": 4550}, {"loss": 1.4308, "grad_norm": 0.745479941368103, "learning_rate": 0.0002, "epoch": 3.5389988358556463, "step": 4560}, {"loss": 1.4229, "grad_norm": 0.7582294940948486, "learning_rate": 0.0002, "epoch": 3.5467597982149788, "step": 4570}, {"loss": 1.4127, "grad_norm": 0.6717444658279419, "learning_rate": 0.0002, "epoch": 3.554520760574311, "step": 4580}, {"loss": 1.4368, "grad_norm": 0.7417883276939392, "learning_rate": 0.0002, "epoch": 3.5622817229336436, "step": 4590}, {"loss": 1.4176, "grad_norm": 0.6385737061500549, "learning_rate": 0.0002, "epoch": 3.570042685292976, "step": 4600}, {"loss": 1.3981, "grad_norm": 0.716704249382019, "learning_rate": 0.0002, "epoch": 3.577803647652309, "step": 4610}, {"loss": 1.3889, "grad_norm": 0.6948980093002319, "learning_rate": 0.0002, "epoch": 3.5855646100116414, "step": 4620}, {"loss": 1.5177, "grad_norm": 0.6961140036582947, "learning_rate": 0.0002, "epoch": 3.593325572370974, "step": 4630}, {"loss": 1.4508, "grad_norm": 0.7493122220039368, "learning_rate": 0.0002, "epoch": 3.6010865347303067, "step": 4640}, {"loss": 1.3987, "grad_norm": 0.7431658506393433, "learning_rate": 0.0002, "epoch": 3.608847497089639, "step": 4650}, {"loss": 1.4551, "grad_norm": 0.8353387713432312, "learning_rate": 0.0002, "epoch": 3.6166084594489716, "step": 4660}, {"loss": 1.4533, "grad_norm": 0.7095612287521362, "learning_rate": 0.0002, "epoch": 3.6243694218083045, "step": 4670}, {"loss": 1.4003, "grad_norm": 0.776620090007782, "learning_rate": 0.0002, "epoch": 3.632130384167637, "step": 4680}, {"loss": 1.4361, "grad_norm": 0.7198925018310547, "learning_rate": 0.0002, "epoch": 3.6398913465269693, "step": 4690}, {"loss": 1.4543, "grad_norm": 0.8238834738731384, "learning_rate": 0.0002, "epoch": 3.6476523088863018, "step": 4700}, {"loss": 1.3958, "grad_norm": 0.6804245710372925, "learning_rate": 0.0002, "epoch": 3.655413271245634, "step": 4710}, {"loss": 1.4158, "grad_norm": 0.8444845676422119, "learning_rate": 0.0002, "epoch": 3.663174233604967, "step": 4720}, {"loss": 1.3825, "grad_norm": 0.743797779083252, "learning_rate": 0.0002, "epoch": 3.6709351959642995, "step": 4730}, {"loss": 1.4213, "grad_norm": 0.8994188904762268, "learning_rate": 0.0002, "epoch": 3.678696158323632, "step": 4740}, {"loss": 1.4281, "grad_norm": 0.75416100025177, "learning_rate": 0.0002, "epoch": 3.686457120682965, "step": 4750}, {"loss": 1.4154, "grad_norm": 0.6499266028404236, "learning_rate": 0.0002, "epoch": 3.6942180830422973, "step": 4760}, {"loss": 1.4005, "grad_norm": 0.7246791124343872, "learning_rate": 0.0002, "epoch": 3.7019790454016297, "step": 4770}, {"loss": 1.426, "grad_norm": 0.7831124067306519, "learning_rate": 0.0002, "epoch": 3.7097400077609626, "step": 4780}, {"loss": 1.3933, "grad_norm": 0.7130028009414673, "learning_rate": 0.0002, "epoch": 3.717500970120295, "step": 4790}, {"loss": 1.4632, "grad_norm": 0.7501602172851562, "learning_rate": 0.0002, "epoch": 3.7252619324796274, "step": 4800}, {"loss": 1.4985, "grad_norm": 0.6980932950973511, "learning_rate": 0.0002, "epoch": 3.73302289483896, "step": 4810}, {"loss": 1.4517, "grad_norm": 0.8050530552864075, "learning_rate": 0.0002, "epoch": 3.7407838571982923, "step": 4820}, {"loss": 1.4703, "grad_norm": 0.6385579705238342, "learning_rate": 0.0002, "epoch": 3.748544819557625, "step": 4830}, {"loss": 1.5281, "grad_norm": 0.6664714813232422, "learning_rate": 0.0002, "epoch": 3.7563057819169576, "step": 4840}, {"loss": 1.4443, "grad_norm": 0.7125676274299622, "learning_rate": 0.0002, "epoch": 3.76406674427629, "step": 4850}, {"loss": 1.3958, "grad_norm": 0.7231866717338562, "learning_rate": 0.0002, "epoch": 3.771827706635623, "step": 4860}, {"loss": 1.4446, "grad_norm": 0.6917183995246887, "learning_rate": 0.0002, "epoch": 3.7795886689949554, "step": 4870}, {"loss": 1.4369, "grad_norm": 0.665037989616394, "learning_rate": 0.0002, "epoch": 3.787349631354288, "step": 4880}, {"loss": 1.4193, "grad_norm": 0.5837726593017578, "learning_rate": 0.0002, "epoch": 3.7951105937136207, "step": 4890}, {"loss": 1.4176, "grad_norm": 0.6366701722145081, "learning_rate": 0.0002, "epoch": 3.802871556072953, "step": 4900}, {"loss": 1.46, "grad_norm": 0.7082223892211914, "learning_rate": 0.0002, "epoch": 3.8106325184322856, "step": 4910}, {"loss": 1.5139, "grad_norm": 0.8101672530174255, "learning_rate": 0.0002, "epoch": 3.818393480791618, "step": 4920}, {"loss": 1.3659, "grad_norm": 0.7516148090362549, "learning_rate": 0.0002, "epoch": 3.826154443150951, "step": 4930}, {"loss": 1.3909, "grad_norm": 0.7928489446640015, "learning_rate": 0.0002, "epoch": 3.8339154055102833, "step": 4940}, {"loss": 1.4255, "grad_norm": 0.6892234683036804, "learning_rate": 0.0002, "epoch": 3.8416763678696157, "step": 4950}, {"loss": 1.5024, "grad_norm": 0.6381304264068604, "learning_rate": 0.0002, "epoch": 3.849437330228948, "step": 4960}, {"loss": 1.4873, "grad_norm": 0.8068831562995911, "learning_rate": 0.0002, "epoch": 3.857198292588281, "step": 4970}, {"loss": 1.45, "grad_norm": 0.7289869785308838, "learning_rate": 0.0002, "epoch": 3.8649592549476135, "step": 4980}, {"loss": 1.398, "grad_norm": 0.7278549075126648, "learning_rate": 0.0002, "epoch": 3.872720217306946, "step": 4990}, {"loss": 1.4442, "grad_norm": 0.7324236631393433, "learning_rate": 0.0002, "epoch": 3.880481179666279, "step": 5000}, {"loss": 1.4511, "grad_norm": 0.6759871244430542, "learning_rate": 0.0002, "epoch": 3.8882421420256112, "step": 5010}, {"loss": 1.4705, "grad_norm": 0.8159207701683044, "learning_rate": 0.0002, "epoch": 3.8960031043849437, "step": 5020}, {"loss": 1.4685, "grad_norm": 0.6536211967468262, "learning_rate": 0.0002, "epoch": 3.9037640667442766, "step": 5030}, {"loss": 1.4335, "grad_norm": 0.6827932000160217, "learning_rate": 0.0002, "epoch": 3.911525029103609, "step": 5040}, {"loss": 1.433, "grad_norm": 0.6688340306282043, "learning_rate": 0.0002, "epoch": 3.9192859914629414, "step": 5050}, {"loss": 1.4099, "grad_norm": 0.6385695934295654, "learning_rate": 0.0002, "epoch": 3.927046953822274, "step": 5060}, {"loss": 1.4767, "grad_norm": 0.6975107192993164, "learning_rate": 0.0002, "epoch": 3.9348079161816063, "step": 5070}, {"loss": 1.4893, "grad_norm": 0.6684112548828125, "learning_rate": 0.0002, "epoch": 3.942568878540939, "step": 5080}, {"loss": 1.4732, "grad_norm": 0.8349628448486328, "learning_rate": 0.0002, "epoch": 3.9503298409002716, "step": 5090}, {"loss": 1.5131, "grad_norm": 0.7146425843238831, "learning_rate": 0.0002, "epoch": 3.958090803259604, "step": 5100}, {"loss": 1.4149, "grad_norm": 0.6555036902427673, "learning_rate": 0.0002, "epoch": 3.965851765618937, "step": 5110}, {"loss": 1.4274, "grad_norm": 0.7037415504455566, "learning_rate": 0.0002, "epoch": 3.9736127279782694, "step": 5120}, {"loss": 1.4292, "grad_norm": 0.7235575914382935, "learning_rate": 0.0002, "epoch": 3.981373690337602, "step": 5130}, {"loss": 1.4455, "grad_norm": 0.7092325687408447, "learning_rate": 0.0002, "epoch": 3.9891346526969347, "step": 5140}, {"loss": 1.4512, "grad_norm": 0.7490319609642029, "learning_rate": 0.0002, "epoch": 3.996895615056267, "step": 5150}]}
+{"epoch": 4.9996119518820334, "step": 6442, "epoch_duration": 3756.667355775833, "total_accumulated_duration": 18030.87024497986, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}, {"eval_loss": 1.8472607135772705, "eval_runtime": 106.5541, "eval_samples_per_second": 4.758, "eval_steps_per_second": 0.601, "epoch": 2.9996119518820334, "step": 3865}, {"loss": 1.4983, "grad_norm": 0.5645653605461121, "learning_rate": 0.0002, "epoch": 3.0034924330616994, "step": 3870}, {"loss": 1.4334, "grad_norm": 0.6457151174545288, "learning_rate": 0.0002, "epoch": 3.0112533954210323, "step": 3880}, {"loss": 1.3899, "grad_norm": 0.583838164806366, "learning_rate": 0.0002, "epoch": 3.0190143577803648, "step": 3890}, {"loss": 1.3258, "grad_norm": 0.6819260120391846, "learning_rate": 0.0002, "epoch": 3.026775320139697, "step": 3900}, {"loss": 1.3458, "grad_norm": 0.6692903637886047, "learning_rate": 0.0002, "epoch": 3.03453628249903, "step": 3910}, {"loss": 1.4356, "grad_norm": 0.6101024746894836, "learning_rate": 0.0002, "epoch": 3.0422972448583625, "step": 3920}, {"loss": 1.394, "grad_norm": 0.7014093399047852, "learning_rate": 0.0002, "epoch": 3.050058207217695, "step": 3930}, {"loss": 1.3885, "grad_norm": 0.7380381226539612, "learning_rate": 0.0002, "epoch": 3.0578191695770274, "step": 3940}, {"loss": 1.4206, "grad_norm": 0.6607900857925415, "learning_rate": 0.0002, "epoch": 3.0655801319363603, "step": 3950}, {"loss": 1.4293, "grad_norm": 0.735263466835022, "learning_rate": 0.0002, "epoch": 3.0733410942956927, "step": 3960}, {"loss": 1.3966, "grad_norm": 0.6788513660430908, "learning_rate": 0.0002, "epoch": 3.081102056655025, "step": 3970}, {"loss": 1.3435, "grad_norm": 0.6347652673721313, "learning_rate": 0.0002, "epoch": 3.088863019014358, "step": 3980}, {"loss": 1.4518, "grad_norm": 0.7056642770767212, "learning_rate": 0.0002, "epoch": 3.0966239813736904, "step": 3990}, {"loss": 1.4474, "grad_norm": 0.6387075185775757, "learning_rate": 0.0002, "epoch": 3.104384943733023, "step": 4000}, {"loss": 1.3833, "grad_norm": 0.6701116561889648, "learning_rate": 0.0002, "epoch": 3.1121459060923553, "step": 4010}, {"loss": 1.404, "grad_norm": 0.7558449506759644, "learning_rate": 0.0002, "epoch": 3.119906868451688, "step": 4020}, {"loss": 1.3294, "grad_norm": 0.6612881422042847, "learning_rate": 0.0002, "epoch": 3.1276678308110206, "step": 4030}, {"loss": 1.439, "grad_norm": 0.7474587559700012, "learning_rate": 0.0002, "epoch": 3.135428793170353, "step": 4040}, {"loss": 1.4616, "grad_norm": 0.7292373776435852, "learning_rate": 0.0002, "epoch": 3.1431897555296855, "step": 4050}, {"loss": 1.3908, "grad_norm": 0.7432886958122253, "learning_rate": 0.0002, "epoch": 3.1509507178890184, "step": 4060}, {"loss": 1.4214, "grad_norm": 0.6366098523139954, "learning_rate": 0.0002, "epoch": 3.158711680248351, "step": 4070}, {"loss": 1.5044, "grad_norm": 0.6837611794471741, "learning_rate": 0.0002, "epoch": 3.1664726426076832, "step": 4080}, {"loss": 1.4332, "grad_norm": 0.7194393277168274, "learning_rate": 0.0002, "epoch": 3.174233604967016, "step": 4090}, {"loss": 1.3628, "grad_norm": 0.6963607668876648, "learning_rate": 0.0002, "epoch": 3.1819945673263486, "step": 4100}, {"loss": 1.4127, "grad_norm": 0.6404902935028076, "learning_rate": 0.0002, "epoch": 3.189755529685681, "step": 4110}, {"loss": 1.4394, "grad_norm": 0.7172070741653442, "learning_rate": 0.0002, "epoch": 3.1975164920450134, "step": 4120}, {"loss": 1.4658, "grad_norm": 0.6577759385108948, "learning_rate": 0.0002, "epoch": 3.2052774544043463, "step": 4130}, {"loss": 1.4019, "grad_norm": 0.6658480167388916, "learning_rate": 0.0002, "epoch": 3.2130384167636787, "step": 4140}, {"loss": 1.4348, "grad_norm": 0.6771699786186218, "learning_rate": 0.0002, "epoch": 3.220799379123011, "step": 4150}, {"loss": 1.4736, "grad_norm": 0.699035108089447, "learning_rate": 0.0002, "epoch": 3.2285603414823436, "step": 4160}, {"loss": 1.4096, "grad_norm": 0.7218514680862427, "learning_rate": 0.0002, "epoch": 3.2363213038416765, "step": 4170}, {"loss": 1.3637, "grad_norm": 0.6270631551742554, "learning_rate": 0.0002, "epoch": 3.244082266201009, "step": 4180}, {"loss": 1.4076, "grad_norm": 0.6828921437263489, "learning_rate": 0.0002, "epoch": 3.2518432285603414, "step": 4190}, {"loss": 1.4663, "grad_norm": 0.6005498170852661, "learning_rate": 0.0002, "epoch": 3.2596041909196742, "step": 4200}, {"loss": 1.4798, "grad_norm": 0.6974790692329407, "learning_rate": 0.0002, "epoch": 3.2673651532790067, "step": 4210}, {"loss": 1.5012, "grad_norm": 0.7269543409347534, "learning_rate": 0.0002, "epoch": 3.275126115638339, "step": 4220}, {"loss": 1.3848, "grad_norm": 0.6728787422180176, "learning_rate": 0.0002, "epoch": 3.2828870779976715, "step": 4230}, {"loss": 1.4112, "grad_norm": 0.676972508430481, "learning_rate": 0.0002, "epoch": 3.2906480403570044, "step": 4240}, {"loss": 1.4206, "grad_norm": 0.748309314250946, "learning_rate": 0.0002, "epoch": 3.298409002716337, "step": 4250}, {"loss": 1.4973, "grad_norm": 0.6976589560508728, "learning_rate": 0.0002, "epoch": 3.3061699650756693, "step": 4260}, {"loss": 1.3967, "grad_norm": 0.649780809879303, "learning_rate": 0.0002, "epoch": 3.3139309274350017, "step": 4270}, {"loss": 1.327, "grad_norm": 0.6529902815818787, "learning_rate": 0.0002, "epoch": 3.3216918897943346, "step": 4280}, {"loss": 1.4888, "grad_norm": 0.9273163676261902, "learning_rate": 0.0002, "epoch": 3.329452852153667, "step": 4290}, {"loss": 1.4859, "grad_norm": 0.717024028301239, "learning_rate": 0.0002, "epoch": 3.3372138145129995, "step": 4300}, {"loss": 1.4441, "grad_norm": 0.7914950251579285, "learning_rate": 0.0002, "epoch": 3.3449747768723324, "step": 4310}, {"loss": 1.432, "grad_norm": 0.7133203148841858, "learning_rate": 0.0002, "epoch": 3.352735739231665, "step": 4320}, {"loss": 1.4662, "grad_norm": 0.7409568428993225, "learning_rate": 0.0002, "epoch": 3.3604967015909972, "step": 4330}, {"loss": 1.3992, "grad_norm": 0.6993981003761292, "learning_rate": 0.0002, "epoch": 3.3682576639503297, "step": 4340}, {"loss": 1.4261, "grad_norm": 0.7114535570144653, "learning_rate": 0.0002, "epoch": 3.3760186263096625, "step": 4350}, {"loss": 1.4227, "grad_norm": 0.6790860295295715, "learning_rate": 0.0002, "epoch": 3.383779588668995, "step": 4360}, {"loss": 1.4128, "grad_norm": 0.6507849097251892, "learning_rate": 0.0002, "epoch": 3.3915405510283274, "step": 4370}, {"loss": 1.4559, "grad_norm": 0.5967804193496704, "learning_rate": 0.0002, "epoch": 3.39930151338766, "step": 4380}, {"loss": 1.3687, "grad_norm": 0.6625847816467285, "learning_rate": 0.0002, "epoch": 3.4070624757469927, "step": 4390}, {"loss": 1.4193, "grad_norm": 0.6736508011817932, "learning_rate": 0.0002, "epoch": 3.414823438106325, "step": 4400}, {"loss": 1.4363, "grad_norm": 0.7870860695838928, "learning_rate": 0.0002, "epoch": 3.4225844004656576, "step": 4410}, {"loss": 1.4114, "grad_norm": 0.7205295562744141, "learning_rate": 0.0002, "epoch": 3.4303453628249905, "step": 4420}, {"loss": 1.4131, "grad_norm": 0.6634634137153625, "learning_rate": 0.0002, "epoch": 3.438106325184323, "step": 4430}, {"loss": 1.4683, "grad_norm": 0.7562733292579651, "learning_rate": 0.0002, "epoch": 3.4458672875436553, "step": 4440}, {"loss": 1.3486, "grad_norm": 0.6585879921913147, "learning_rate": 0.0002, "epoch": 3.453628249902988, "step": 4450}, {"loss": 1.4283, "grad_norm": 0.6896792054176331, "learning_rate": 0.0002, "epoch": 3.4613892122623207, "step": 4460}, {"loss": 1.4208, "grad_norm": 0.6520342230796814, "learning_rate": 0.0002, "epoch": 3.469150174621653, "step": 4470}, {"loss": 1.3423, "grad_norm": 0.6760806441307068, "learning_rate": 0.0002, "epoch": 3.4769111369809855, "step": 4480}, {"loss": 1.4398, "grad_norm": 0.7539774179458618, "learning_rate": 0.0002, "epoch": 3.484672099340318, "step": 4490}, {"loss": 1.4534, "grad_norm": 0.7409411668777466, "learning_rate": 0.0002, "epoch": 3.492433061699651, "step": 4500}, {"loss": 1.4069, "grad_norm": 0.6876253485679626, "learning_rate": 0.0002, "epoch": 3.5001940240589833, "step": 4510}, {"loss": 1.4228, "grad_norm": 0.7028461694717407, "learning_rate": 0.0002, "epoch": 3.5079549864183157, "step": 4520}, {"loss": 1.4723, "grad_norm": 0.8056529760360718, "learning_rate": 0.0002, "epoch": 3.5157159487776486, "step": 4530}, {"loss": 1.4148, "grad_norm": 0.711338996887207, "learning_rate": 0.0002, "epoch": 3.523476911136981, "step": 4540}, {"loss": 1.5247, "grad_norm": 0.7343552708625793, "learning_rate": 0.0002, "epoch": 3.5312378734963135, "step": 4550}, {"loss": 1.4308, "grad_norm": 0.745479941368103, "learning_rate": 0.0002, "epoch": 3.5389988358556463, "step": 4560}, {"loss": 1.4229, "grad_norm": 0.7582294940948486, "learning_rate": 0.0002, "epoch": 3.5467597982149788, "step": 4570}, {"loss": 1.4127, "grad_norm": 0.6717444658279419, "learning_rate": 0.0002, "epoch": 3.554520760574311, "step": 4580}, {"loss": 1.4368, "grad_norm": 0.7417883276939392, "learning_rate": 0.0002, "epoch": 3.5622817229336436, "step": 4590}, {"loss": 1.4176, "grad_norm": 0.6385737061500549, "learning_rate": 0.0002, "epoch": 3.570042685292976, "step": 4600}, {"loss": 1.3981, "grad_norm": 0.716704249382019, "learning_rate": 0.0002, "epoch": 3.577803647652309, "step": 4610}, {"loss": 1.3889, "grad_norm": 0.6948980093002319, "learning_rate": 0.0002, "epoch": 3.5855646100116414, "step": 4620}, {"loss": 1.5177, "grad_norm": 0.6961140036582947, "learning_rate": 0.0002, "epoch": 3.593325572370974, "step": 4630}, {"loss": 1.4508, "grad_norm": 0.7493122220039368, "learning_rate": 0.0002, "epoch": 3.6010865347303067, "step": 4640}, {"loss": 1.3987, "grad_norm": 0.7431658506393433, "learning_rate": 0.0002, "epoch": 3.608847497089639, "step": 4650}, {"loss": 1.4551, "grad_norm": 0.8353387713432312, "learning_rate": 0.0002, "epoch": 3.6166084594489716, "step": 4660}, {"loss": 1.4533, "grad_norm": 0.7095612287521362, "learning_rate": 0.0002, "epoch": 3.6243694218083045, "step": 4670}, {"loss": 1.4003, "grad_norm": 0.776620090007782, "learning_rate": 0.0002, "epoch": 3.632130384167637, "step": 4680}, {"loss": 1.4361, "grad_norm": 0.7198925018310547, "learning_rate": 0.0002, "epoch": 3.6398913465269693, "step": 4690}, {"loss": 1.4543, "grad_norm": 0.8238834738731384, "learning_rate": 0.0002, "epoch": 3.6476523088863018, "step": 4700}, {"loss": 1.3958, "grad_norm": 0.6804245710372925, "learning_rate": 0.0002, "epoch": 3.655413271245634, "step": 4710}, {"loss": 1.4158, "grad_norm": 0.8444845676422119, "learning_rate": 0.0002, "epoch": 3.663174233604967, "step": 4720}, {"loss": 1.3825, "grad_norm": 0.743797779083252, "learning_rate": 0.0002, "epoch": 3.6709351959642995, "step": 4730}, {"loss": 1.4213, "grad_norm": 0.8994188904762268, "learning_rate": 0.0002, "epoch": 3.678696158323632, "step": 4740}, {"loss": 1.4281, "grad_norm": 0.75416100025177, "learning_rate": 0.0002, "epoch": 3.686457120682965, "step": 4750}, {"loss": 1.4154, "grad_norm": 0.6499266028404236, "learning_rate": 0.0002, "epoch": 3.6942180830422973, "step": 4760}, {"loss": 1.4005, "grad_norm": 0.7246791124343872, "learning_rate": 0.0002, "epoch": 3.7019790454016297, "step": 4770}, {"loss": 1.426, "grad_norm": 0.7831124067306519, "learning_rate": 0.0002, "epoch": 3.7097400077609626, "step": 4780}, {"loss": 1.3933, "grad_norm": 0.7130028009414673, "learning_rate": 0.0002, "epoch": 3.717500970120295, "step": 4790}, {"loss": 1.4632, "grad_norm": 0.7501602172851562, "learning_rate": 0.0002, "epoch": 3.7252619324796274, "step": 4800}, {"loss": 1.4985, "grad_norm": 0.6980932950973511, "learning_rate": 0.0002, "epoch": 3.73302289483896, "step": 4810}, {"loss": 1.4517, "grad_norm": 0.8050530552864075, "learning_rate": 0.0002, "epoch": 3.7407838571982923, "step": 4820}, {"loss": 1.4703, "grad_norm": 0.6385579705238342, "learning_rate": 0.0002, "epoch": 3.748544819557625, "step": 4830}, {"loss": 1.5281, "grad_norm": 0.6664714813232422, "learning_rate": 0.0002, "epoch": 3.7563057819169576, "step": 4840}, {"loss": 1.4443, "grad_norm": 0.7125676274299622, "learning_rate": 0.0002, "epoch": 3.76406674427629, "step": 4850}, {"loss": 1.3958, "grad_norm": 0.7231866717338562, "learning_rate": 0.0002, "epoch": 3.771827706635623, "step": 4860}, {"loss": 1.4446, "grad_norm": 0.6917183995246887, "learning_rate": 0.0002, "epoch": 3.7795886689949554, "step": 4870}, {"loss": 1.4369, "grad_norm": 0.665037989616394, "learning_rate": 0.0002, "epoch": 3.787349631354288, "step": 4880}, {"loss": 1.4193, "grad_norm": 0.5837726593017578, "learning_rate": 0.0002, "epoch": 3.7951105937136207, "step": 4890}, {"loss": 1.4176, "grad_norm": 0.6366701722145081, "learning_rate": 0.0002, "epoch": 3.802871556072953, "step": 4900}, {"loss": 1.46, "grad_norm": 0.7082223892211914, "learning_rate": 0.0002, "epoch": 3.8106325184322856, "step": 4910}, {"loss": 1.5139, "grad_norm": 0.8101672530174255, "learning_rate": 0.0002, "epoch": 3.818393480791618, "step": 4920}, {"loss": 1.3659, "grad_norm": 0.7516148090362549, "learning_rate": 0.0002, "epoch": 3.826154443150951, "step": 4930}, {"loss": 1.3909, "grad_norm": 0.7928489446640015, "learning_rate": 0.0002, "epoch": 3.8339154055102833, "step": 4940}, {"loss": 1.4255, "grad_norm": 0.6892234683036804, "learning_rate": 0.0002, "epoch": 3.8416763678696157, "step": 4950}, {"loss": 1.5024, "grad_norm": 0.6381304264068604, "learning_rate": 0.0002, "epoch": 3.849437330228948, "step": 4960}, {"loss": 1.4873, "grad_norm": 0.8068831562995911, "learning_rate": 0.0002, "epoch": 3.857198292588281, "step": 4970}, {"loss": 1.45, "grad_norm": 0.7289869785308838, "learning_rate": 0.0002, "epoch": 3.8649592549476135, "step": 4980}, {"loss": 1.398, "grad_norm": 0.7278549075126648, "learning_rate": 0.0002, "epoch": 3.872720217306946, "step": 4990}, {"loss": 1.4442, "grad_norm": 0.7324236631393433, "learning_rate": 0.0002, "epoch": 3.880481179666279, "step": 5000}, {"loss": 1.4511, "grad_norm": 0.6759871244430542, "learning_rate": 0.0002, "epoch": 3.8882421420256112, "step": 5010}, {"loss": 1.4705, "grad_norm": 0.8159207701683044, "learning_rate": 0.0002, "epoch": 3.8960031043849437, "step": 5020}, {"loss": 1.4685, "grad_norm": 0.6536211967468262, "learning_rate": 0.0002, "epoch": 3.9037640667442766, "step": 5030}, {"loss": 1.4335, "grad_norm": 0.6827932000160217, "learning_rate": 0.0002, "epoch": 3.911525029103609, "step": 5040}, {"loss": 1.433, "grad_norm": 0.6688340306282043, "learning_rate": 0.0002, "epoch": 3.9192859914629414, "step": 5050}, {"loss": 1.4099, "grad_norm": 0.6385695934295654, "learning_rate": 0.0002, "epoch": 3.927046953822274, "step": 5060}, {"loss": 1.4767, "grad_norm": 0.6975107192993164, "learning_rate": 0.0002, "epoch": 3.9348079161816063, "step": 5070}, {"loss": 1.4893, "grad_norm": 0.6684112548828125, "learning_rate": 0.0002, "epoch": 3.942568878540939, "step": 5080}, {"loss": 1.4732, "grad_norm": 0.8349628448486328, "learning_rate": 0.0002, "epoch": 3.9503298409002716, "step": 5090}, {"loss": 1.5131, "grad_norm": 0.7146425843238831, "learning_rate": 0.0002, "epoch": 3.958090803259604, "step": 5100}, {"loss": 1.4149, "grad_norm": 0.6555036902427673, "learning_rate": 0.0002, "epoch": 3.965851765618937, "step": 5110}, {"loss": 1.4274, "grad_norm": 0.7037415504455566, "learning_rate": 0.0002, "epoch": 3.9736127279782694, "step": 5120}, {"loss": 1.4292, "grad_norm": 0.7235575914382935, "learning_rate": 0.0002, "epoch": 3.981373690337602, "step": 5130}, {"loss": 1.4455, "grad_norm": 0.7092325687408447, "learning_rate": 0.0002, "epoch": 3.9891346526969347, "step": 5140}, {"loss": 1.4512, "grad_norm": 0.7490319609642029, "learning_rate": 0.0002, "epoch": 3.996895615056267, "step": 5150}, {"eval_loss": 1.9131355285644531, "eval_runtime": 105.5778, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 4.0, "step": 5154}, {"loss": 1.2643, "grad_norm": 0.7075854539871216, "learning_rate": 0.0002, "epoch": 4.0046565774155995, "step": 5160}, {"loss": 1.209, "grad_norm": 0.9466007351875305, "learning_rate": 0.0002, "epoch": 4.012417539774932, "step": 5170}, {"loss": 1.2567, "grad_norm": 1.0297044515609741, "learning_rate": 0.0002, "epoch": 4.020178502134264, "step": 5180}, {"loss": 1.1796, "grad_norm": 0.7765059471130371, "learning_rate": 0.0002, "epoch": 4.027939464493597, "step": 5190}, {"loss": 1.2356, "grad_norm": 0.995760977268219, "learning_rate": 0.0002, "epoch": 4.03570042685293, "step": 5200}, {"loss": 1.1792, "grad_norm": 0.8663829565048218, "learning_rate": 0.0002, "epoch": 4.043461389212262, "step": 5210}, {"loss": 1.2471, "grad_norm": 1.0660825967788696, "learning_rate": 0.0002, "epoch": 4.051222351571595, "step": 5220}, {"loss": 1.1676, "grad_norm": 0.9858174920082092, "learning_rate": 0.0002, "epoch": 4.058983313930927, "step": 5230}, {"loss": 1.2448, "grad_norm": 0.8911338448524475, "learning_rate": 0.0002, "epoch": 4.06674427629026, "step": 5240}, {"loss": 1.1858, "grad_norm": 1.0848394632339478, "learning_rate": 0.0002, "epoch": 4.074505238649593, "step": 5250}, {"loss": 1.1684, "grad_norm": 1.0849905014038086, "learning_rate": 0.0002, "epoch": 4.082266201008925, "step": 5260}, {"loss": 1.2007, "grad_norm": 1.0497841835021973, "learning_rate": 0.0002, "epoch": 4.090027163368258, "step": 5270}, {"loss": 1.2552, "grad_norm": 0.8943053483963013, "learning_rate": 0.0002, "epoch": 4.0977881257275905, "step": 5280}, {"loss": 1.1923, "grad_norm": 0.8432527184486389, "learning_rate": 0.0002, "epoch": 4.1055490880869225, "step": 5290}, {"loss": 1.1634, "grad_norm": 0.9690414667129517, "learning_rate": 0.0002, "epoch": 4.113310050446255, "step": 5300}, {"loss": 1.3019, "grad_norm": 0.7790773510932922, "learning_rate": 0.0002, "epoch": 4.121071012805588, "step": 5310}, {"loss": 1.1806, "grad_norm": 0.9289211630821228, "learning_rate": 0.0002, "epoch": 4.12883197516492, "step": 5320}, {"loss": 1.1458, "grad_norm": 1.0785125494003296, "learning_rate": 0.0002, "epoch": 4.136592937524253, "step": 5330}, {"loss": 1.2086, "grad_norm": 0.8559591770172119, "learning_rate": 0.0002, "epoch": 4.144353899883585, "step": 5340}, {"loss": 1.1974, "grad_norm": 0.9405956268310547, "learning_rate": 0.0002, "epoch": 4.152114862242918, "step": 5350}, {"loss": 1.1793, "grad_norm": 0.9942827820777893, "learning_rate": 0.0002, "epoch": 4.159875824602251, "step": 5360}, {"loss": 1.1659, "grad_norm": 0.9141933917999268, "learning_rate": 0.0002, "epoch": 4.167636786961583, "step": 5370}, {"loss": 1.1647, "grad_norm": 0.8206015229225159, "learning_rate": 0.0002, "epoch": 4.175397749320916, "step": 5380}, {"loss": 1.2778, "grad_norm": 0.9340888857841492, "learning_rate": 0.0002, "epoch": 4.183158711680249, "step": 5390}, {"loss": 1.2459, "grad_norm": 1.2122114896774292, "learning_rate": 0.0002, "epoch": 4.190919674039581, "step": 5400}, {"loss": 1.2371, "grad_norm": 1.0661298036575317, "learning_rate": 0.0002, "epoch": 4.1986806363989135, "step": 5410}, {"loss": 1.1978, "grad_norm": 0.9372861385345459, "learning_rate": 0.0002, "epoch": 4.206441598758246, "step": 5420}, {"loss": 1.2653, "grad_norm": 0.894012987613678, "learning_rate": 0.0002, "epoch": 4.214202561117578, "step": 5430}, {"loss": 1.387, "grad_norm": 1.0647753477096558, "learning_rate": 0.0002, "epoch": 4.221963523476911, "step": 5440}, {"loss": 1.2231, "grad_norm": 0.989179790019989, "learning_rate": 0.0002, "epoch": 4.229724485836243, "step": 5450}, {"loss": 1.2715, "grad_norm": 1.1601181030273438, "learning_rate": 0.0002, "epoch": 4.237485448195576, "step": 5460}, {"loss": 1.2406, "grad_norm": 0.9395585656166077, "learning_rate": 0.0002, "epoch": 4.245246410554909, "step": 5470}, {"loss": 1.2779, "grad_norm": 0.9527766108512878, "learning_rate": 0.0002, "epoch": 4.253007372914241, "step": 5480}, {"loss": 1.267, "grad_norm": 1.0319520235061646, "learning_rate": 0.0002, "epoch": 4.260768335273574, "step": 5490}, {"loss": 1.2633, "grad_norm": 0.8659824728965759, "learning_rate": 0.0002, "epoch": 4.268529297632907, "step": 5500}, {"loss": 1.1475, "grad_norm": 1.099211573600769, "learning_rate": 0.0002, "epoch": 4.276290259992239, "step": 5510}, {"loss": 1.2508, "grad_norm": 0.9363361597061157, "learning_rate": 0.0002, "epoch": 4.284051222351572, "step": 5520}, {"loss": 1.189, "grad_norm": 0.8437647223472595, "learning_rate": 0.0002, "epoch": 4.2918121847109045, "step": 5530}, {"loss": 1.2212, "grad_norm": 0.9181258678436279, "learning_rate": 0.0002, "epoch": 4.2995731470702365, "step": 5540}, {"loss": 1.2092, "grad_norm": 0.9059357643127441, "learning_rate": 0.0002, "epoch": 4.307334109429569, "step": 5550}, {"loss": 1.2189, "grad_norm": 0.9337241649627686, "learning_rate": 0.0002, "epoch": 4.315095071788901, "step": 5560}, {"loss": 1.2462, "grad_norm": 0.9428889155387878, "learning_rate": 0.0002, "epoch": 4.322856034148234, "step": 5570}, {"loss": 1.2675, "grad_norm": 1.003589153289795, "learning_rate": 0.0002, "epoch": 4.330616996507567, "step": 5580}, {"loss": 1.2703, "grad_norm": 1.1249268054962158, "learning_rate": 0.0002, "epoch": 4.338377958866899, "step": 5590}, {"loss": 1.2501, "grad_norm": 0.8623469471931458, "learning_rate": 0.0002, "epoch": 4.346138921226232, "step": 5600}, {"loss": 1.2404, "grad_norm": 1.1389174461364746, "learning_rate": 0.0002, "epoch": 4.353899883585565, "step": 5610}, {"loss": 1.2245, "grad_norm": 1.0136264562606812, "learning_rate": 0.0002, "epoch": 4.361660845944897, "step": 5620}, {"loss": 1.3473, "grad_norm": 0.9567070603370667, "learning_rate": 0.0002, "epoch": 4.36942180830423, "step": 5630}, {"loss": 1.2988, "grad_norm": 1.0592148303985596, "learning_rate": 0.0002, "epoch": 4.377182770663563, "step": 5640}, {"loss": 1.212, "grad_norm": 1.0110485553741455, "learning_rate": 0.0002, "epoch": 4.384943733022895, "step": 5650}, {"loss": 1.2086, "grad_norm": 0.9914907217025757, "learning_rate": 0.0002, "epoch": 4.3927046953822275, "step": 5660}, {"loss": 1.2363, "grad_norm": 0.9447247982025146, "learning_rate": 0.0002, "epoch": 4.4004656577415595, "step": 5670}, {"loss": 1.2617, "grad_norm": 0.9644378423690796, "learning_rate": 0.0002, "epoch": 4.408226620100892, "step": 5680}, {"loss": 1.2773, "grad_norm": 0.920676589012146, "learning_rate": 0.0002, "epoch": 4.415987582460225, "step": 5690}, {"loss": 1.2792, "grad_norm": 1.060570478439331, "learning_rate": 0.0002, "epoch": 4.423748544819557, "step": 5700}, {"loss": 1.2374, "grad_norm": 0.8857738971710205, "learning_rate": 0.0002, "epoch": 4.43150950717889, "step": 5710}, {"loss": 1.2588, "grad_norm": 1.0536398887634277, "learning_rate": 0.0002, "epoch": 4.439270469538223, "step": 5720}, {"loss": 1.2051, "grad_norm": 0.990847110748291, "learning_rate": 0.0002, "epoch": 4.447031431897555, "step": 5730}, {"loss": 1.2469, "grad_norm": 0.9692499041557312, "learning_rate": 0.0002, "epoch": 4.454792394256888, "step": 5740}, {"loss": 1.2269, "grad_norm": 1.0376402139663696, "learning_rate": 0.0002, "epoch": 4.462553356616221, "step": 5750}, {"loss": 1.1701, "grad_norm": 1.3863259553909302, "learning_rate": 0.0002, "epoch": 4.470314318975553, "step": 5760}, {"loss": 1.2591, "grad_norm": 0.978379487991333, "learning_rate": 0.0002, "epoch": 4.478075281334886, "step": 5770}, {"loss": 1.2729, "grad_norm": 1.0973085165023804, "learning_rate": 0.0002, "epoch": 4.485836243694218, "step": 5780}, {"loss": 1.2404, "grad_norm": 1.057006597518921, "learning_rate": 0.0002, "epoch": 4.4935972060535505, "step": 5790}, {"loss": 1.2476, "grad_norm": 0.9247729182243347, "learning_rate": 0.0002, "epoch": 4.501358168412883, "step": 5800}, {"loss": 1.2369, "grad_norm": 1.0447787046432495, "learning_rate": 0.0002, "epoch": 4.509119130772215, "step": 5810}, {"loss": 1.211, "grad_norm": 1.1930429935455322, "learning_rate": 0.0002, "epoch": 4.516880093131548, "step": 5820}, {"loss": 1.2596, "grad_norm": 0.9867590069770813, "learning_rate": 0.0002, "epoch": 4.524641055490881, "step": 5830}, {"loss": 1.2766, "grad_norm": 0.9591100215911865, "learning_rate": 0.0002, "epoch": 4.532402017850213, "step": 5840}, {"loss": 1.2154, "grad_norm": 0.9950753450393677, "learning_rate": 0.0002, "epoch": 4.540162980209546, "step": 5850}, {"loss": 1.2149, "grad_norm": 1.0087506771087646, "learning_rate": 0.0002, "epoch": 4.547923942568879, "step": 5860}, {"loss": 1.3165, "grad_norm": 1.0934417247772217, "learning_rate": 0.0002, "epoch": 4.555684904928211, "step": 5870}, {"loss": 1.3059, "grad_norm": 1.107987403869629, "learning_rate": 0.0002, "epoch": 4.563445867287544, "step": 5880}, {"loss": 1.2184, "grad_norm": 0.9147276878356934, "learning_rate": 0.0002, "epoch": 4.571206829646876, "step": 5890}, {"loss": 1.24, "grad_norm": 1.036780595779419, "learning_rate": 0.0002, "epoch": 4.578967792006209, "step": 5900}, {"loss": 1.2209, "grad_norm": 0.9284719824790955, "learning_rate": 0.0002, "epoch": 4.5867287543655415, "step": 5910}, {"loss": 1.3693, "grad_norm": 0.9141898155212402, "learning_rate": 0.0002, "epoch": 4.5944897167248735, "step": 5920}, {"loss": 1.2319, "grad_norm": 1.0447357892990112, "learning_rate": 0.0002, "epoch": 4.602250679084206, "step": 5930}, {"loss": 1.2667, "grad_norm": 0.9309114217758179, "learning_rate": 0.0002, "epoch": 4.610011641443539, "step": 5940}, {"loss": 1.2827, "grad_norm": 1.2986129522323608, "learning_rate": 0.0002, "epoch": 4.617772603802871, "step": 5950}, {"loss": 1.312, "grad_norm": 0.9221704602241516, "learning_rate": 0.0002, "epoch": 4.625533566162204, "step": 5960}, {"loss": 1.2769, "grad_norm": 0.9228187799453735, "learning_rate": 0.0002, "epoch": 4.633294528521537, "step": 5970}, {"loss": 1.2953, "grad_norm": 0.9483116269111633, "learning_rate": 0.0002, "epoch": 4.641055490880869, "step": 5980}, {"loss": 1.3437, "grad_norm": 1.0218974351882935, "learning_rate": 0.0002, "epoch": 4.648816453240202, "step": 5990}, {"loss": 1.3085, "grad_norm": 0.9764600396156311, "learning_rate": 0.0002, "epoch": 4.656577415599534, "step": 6000}, {"loss": 1.197, "grad_norm": 0.9115710258483887, "learning_rate": 0.0002, "epoch": 4.664338377958867, "step": 6010}, {"loss": 1.1917, "grad_norm": 0.9245651364326477, "learning_rate": 0.0002, "epoch": 4.6720993403182, "step": 6020}, {"loss": 1.2969, "grad_norm": 0.9686311483383179, "learning_rate": 0.0002, "epoch": 4.6798603026775325, "step": 6030}, {"loss": 1.2702, "grad_norm": 1.1807392835617065, "learning_rate": 0.0002, "epoch": 4.6876212650368645, "step": 6040}, {"loss": 1.328, "grad_norm": 1.0358641147613525, "learning_rate": 0.0002, "epoch": 4.695382227396197, "step": 6050}, {"loss": 1.3281, "grad_norm": 0.987332284450531, "learning_rate": 0.0002, "epoch": 4.703143189755529, "step": 6060}, {"loss": 1.2514, "grad_norm": 1.0526494979858398, "learning_rate": 0.0002, "epoch": 4.710904152114862, "step": 6070}, {"loss": 1.2246, "grad_norm": 1.0276758670806885, "learning_rate": 0.0002, "epoch": 4.718665114474195, "step": 6080}, {"loss": 1.3367, "grad_norm": 0.9904406666755676, "learning_rate": 0.0002, "epoch": 4.726426076833527, "step": 6090}, {"loss": 1.2797, "grad_norm": 1.0084882974624634, "learning_rate": 0.0002, "epoch": 4.73418703919286, "step": 6100}, {"loss": 1.2656, "grad_norm": 0.8646450638771057, "learning_rate": 0.0002, "epoch": 4.741948001552192, "step": 6110}, {"loss": 1.3063, "grad_norm": 0.9233377575874329, "learning_rate": 0.0002, "epoch": 4.749708963911525, "step": 6120}, {"loss": 1.2642, "grad_norm": 0.9675140976905823, "learning_rate": 0.0002, "epoch": 4.757469926270858, "step": 6130}, {"loss": 1.3367, "grad_norm": 0.9639796018600464, "learning_rate": 0.0002, "epoch": 4.765230888630191, "step": 6140}, {"loss": 1.276, "grad_norm": 0.925199568271637, "learning_rate": 0.0002, "epoch": 4.772991850989523, "step": 6150}, {"loss": 1.2441, "grad_norm": 1.050901174545288, "learning_rate": 0.0002, "epoch": 4.7807528133488555, "step": 6160}, {"loss": 1.301, "grad_norm": 0.8920623660087585, "learning_rate": 0.0002, "epoch": 4.7885137757081875, "step": 6170}, {"loss": 1.263, "grad_norm": 0.8964757919311523, "learning_rate": 0.0002, "epoch": 4.79627473806752, "step": 6180}, {"loss": 1.2787, "grad_norm": 1.0839070081710815, "learning_rate": 0.0002, "epoch": 4.804035700426853, "step": 6190}, {"loss": 1.2664, "grad_norm": 0.8809942007064819, "learning_rate": 0.0002, "epoch": 4.811796662786185, "step": 6200}, {"loss": 1.321, "grad_norm": 1.0216195583343506, "learning_rate": 0.0002, "epoch": 4.819557625145518, "step": 6210}, {"loss": 1.3033, "grad_norm": 0.892005980014801, "learning_rate": 0.0002, "epoch": 4.827318587504851, "step": 6220}, {"loss": 1.2602, "grad_norm": 0.9957166910171509, "learning_rate": 0.0002, "epoch": 4.835079549864183, "step": 6230}, {"loss": 1.3562, "grad_norm": 0.9720533490180969, "learning_rate": 0.0002, "epoch": 4.842840512223516, "step": 6240}, {"loss": 1.2651, "grad_norm": 0.9336182475090027, "learning_rate": 0.0002, "epoch": 4.850601474582849, "step": 6250}, {"loss": 1.3136, "grad_norm": 1.2611457109451294, "learning_rate": 0.0002, "epoch": 4.858362436942181, "step": 6260}, {"loss": 1.2234, "grad_norm": 0.8927203416824341, "learning_rate": 0.0002, "epoch": 4.866123399301514, "step": 6270}, {"loss": 1.3463, "grad_norm": 0.9706710577011108, "learning_rate": 0.0002, "epoch": 4.873884361660846, "step": 6280}, {"loss": 1.3209, "grad_norm": 1.1461690664291382, "learning_rate": 0.0002, "epoch": 4.8816453240201785, "step": 6290}, {"loss": 1.2566, "grad_norm": 0.9930381178855896, "learning_rate": 0.0002, "epoch": 4.889406286379511, "step": 6300}, {"loss": 1.2568, "grad_norm": 0.91451096534729, "learning_rate": 0.0002, "epoch": 4.897167248738843, "step": 6310}, {"loss": 1.2836, "grad_norm": 1.0319571495056152, "learning_rate": 0.0002, "epoch": 4.904928211098176, "step": 6320}, {"loss": 1.2908, "grad_norm": 0.990140438079834, "learning_rate": 0.0002, "epoch": 4.912689173457509, "step": 6330}, {"loss": 1.3299, "grad_norm": 1.2466117143630981, "learning_rate": 0.0002, "epoch": 4.920450135816841, "step": 6340}, {"loss": 1.2659, "grad_norm": 1.0316979885101318, "learning_rate": 0.0002, "epoch": 4.928211098176174, "step": 6350}, {"loss": 1.3292, "grad_norm": 1.0643759965896606, "learning_rate": 0.0002, "epoch": 4.935972060535507, "step": 6360}, {"loss": 1.2559, "grad_norm": 0.9703279733657837, "learning_rate": 0.0002, "epoch": 4.943733022894839, "step": 6370}, {"loss": 1.2155, "grad_norm": 0.9767927527427673, "learning_rate": 0.0002, "epoch": 4.951493985254172, "step": 6380}, {"loss": 1.2437, "grad_norm": 0.960854172706604, "learning_rate": 0.0002, "epoch": 4.959254947613504, "step": 6390}, {"loss": 1.3314, "grad_norm": 0.9922910332679749, "learning_rate": 0.0002, "epoch": 4.967015909972837, "step": 6400}, {"loss": 1.3018, "grad_norm": 0.956470787525177, "learning_rate": 0.0002, "epoch": 4.9747768723321695, "step": 6410}, {"loss": 1.2794, "grad_norm": 0.9637242555618286, "learning_rate": 0.0002, "epoch": 4.9825378346915015, "step": 6420}, {"loss": 1.3236, "grad_norm": 1.0855202674865723, "learning_rate": 0.0002, "epoch": 4.990298797050834, "step": 6430}, {"loss": 1.3015, "grad_norm": 0.9655316472053528, "learning_rate": 0.0002, "epoch": 4.998059759410167, "step": 6440}]}
+{"epoch": 6.0, "step": 7731, "epoch_duration": 3940.1947734355927, "total_accumulated_duration": 21971.06501841545, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}, {"eval_loss": 1.8472607135772705, "eval_runtime": 106.5541, "eval_samples_per_second": 4.758, "eval_steps_per_second": 0.601, "epoch": 2.9996119518820334, "step": 3865}, {"loss": 1.4983, "grad_norm": 0.5645653605461121, "learning_rate": 0.0002, "epoch": 3.0034924330616994, "step": 3870}, {"loss": 1.4334, "grad_norm": 0.6457151174545288, "learning_rate": 0.0002, "epoch": 3.0112533954210323, "step": 3880}, {"loss": 1.3899, "grad_norm": 0.583838164806366, "learning_rate": 0.0002, "epoch": 3.0190143577803648, "step": 3890}, {"loss": 1.3258, "grad_norm": 0.6819260120391846, "learning_rate": 0.0002, "epoch": 3.026775320139697, "step": 3900}, {"loss": 1.3458, "grad_norm": 0.6692903637886047, "learning_rate": 0.0002, "epoch": 3.03453628249903, "step": 3910}, {"loss": 1.4356, "grad_norm": 0.6101024746894836, "learning_rate": 0.0002, "epoch": 3.0422972448583625, "step": 3920}, {"loss": 1.394, "grad_norm": 0.7014093399047852, "learning_rate": 0.0002, "epoch": 3.050058207217695, "step": 3930}, {"loss": 1.3885, "grad_norm": 0.7380381226539612, "learning_rate": 0.0002, "epoch": 3.0578191695770274, "step": 3940}, {"loss": 1.4206, "grad_norm": 0.6607900857925415, "learning_rate": 0.0002, "epoch": 3.0655801319363603, "step": 3950}, {"loss": 1.4293, "grad_norm": 0.735263466835022, "learning_rate": 0.0002, "epoch": 3.0733410942956927, "step": 3960}, {"loss": 1.3966, "grad_norm": 0.6788513660430908, "learning_rate": 0.0002, "epoch": 3.081102056655025, "step": 3970}, {"loss": 1.3435, "grad_norm": 0.6347652673721313, "learning_rate": 0.0002, "epoch": 3.088863019014358, "step": 3980}, {"loss": 1.4518, "grad_norm": 0.7056642770767212, "learning_rate": 0.0002, "epoch": 3.0966239813736904, "step": 3990}, {"loss": 1.4474, "grad_norm": 0.6387075185775757, "learning_rate": 0.0002, "epoch": 3.104384943733023, "step": 4000}, {"loss": 1.3833, "grad_norm": 0.6701116561889648, "learning_rate": 0.0002, "epoch": 3.1121459060923553, "step": 4010}, {"loss": 1.404, "grad_norm": 0.7558449506759644, "learning_rate": 0.0002, "epoch": 3.119906868451688, "step": 4020}, {"loss": 1.3294, "grad_norm": 0.6612881422042847, "learning_rate": 0.0002, "epoch": 3.1276678308110206, "step": 4030}, {"loss": 1.439, "grad_norm": 0.7474587559700012, "learning_rate": 0.0002, "epoch": 3.135428793170353, "step": 4040}, {"loss": 1.4616, "grad_norm": 0.7292373776435852, "learning_rate": 0.0002, "epoch": 3.1431897555296855, "step": 4050}, {"loss": 1.3908, "grad_norm": 0.7432886958122253, "learning_rate": 0.0002, "epoch": 3.1509507178890184, "step": 4060}, {"loss": 1.4214, "grad_norm": 0.6366098523139954, "learning_rate": 0.0002, "epoch": 3.158711680248351, "step": 4070}, {"loss": 1.5044, "grad_norm": 0.6837611794471741, "learning_rate": 0.0002, "epoch": 3.1664726426076832, "step": 4080}, {"loss": 1.4332, "grad_norm": 0.7194393277168274, "learning_rate": 0.0002, "epoch": 3.174233604967016, "step": 4090}, {"loss": 1.3628, "grad_norm": 0.6963607668876648, "learning_rate": 0.0002, "epoch": 3.1819945673263486, "step": 4100}, {"loss": 1.4127, "grad_norm": 0.6404902935028076, "learning_rate": 0.0002, "epoch": 3.189755529685681, "step": 4110}, {"loss": 1.4394, "grad_norm": 0.7172070741653442, "learning_rate": 0.0002, "epoch": 3.1975164920450134, "step": 4120}, {"loss": 1.4658, "grad_norm": 0.6577759385108948, "learning_rate": 0.0002, "epoch": 3.2052774544043463, "step": 4130}, {"loss": 1.4019, "grad_norm": 0.6658480167388916, "learning_rate": 0.0002, "epoch": 3.2130384167636787, "step": 4140}, {"loss": 1.4348, "grad_norm": 0.6771699786186218, "learning_rate": 0.0002, "epoch": 3.220799379123011, "step": 4150}, {"loss": 1.4736, "grad_norm": 0.699035108089447, "learning_rate": 0.0002, "epoch": 3.2285603414823436, "step": 4160}, {"loss": 1.4096, "grad_norm": 0.7218514680862427, "learning_rate": 0.0002, "epoch": 3.2363213038416765, "step": 4170}, {"loss": 1.3637, "grad_norm": 0.6270631551742554, "learning_rate": 0.0002, "epoch": 3.244082266201009, "step": 4180}, {"loss": 1.4076, "grad_norm": 0.6828921437263489, "learning_rate": 0.0002, "epoch": 3.2518432285603414, "step": 4190}, {"loss": 1.4663, "grad_norm": 0.6005498170852661, "learning_rate": 0.0002, "epoch": 3.2596041909196742, "step": 4200}, {"loss": 1.4798, "grad_norm": 0.6974790692329407, "learning_rate": 0.0002, "epoch": 3.2673651532790067, "step": 4210}, {"loss": 1.5012, "grad_norm": 0.7269543409347534, "learning_rate": 0.0002, "epoch": 3.275126115638339, "step": 4220}, {"loss": 1.3848, "grad_norm": 0.6728787422180176, "learning_rate": 0.0002, "epoch": 3.2828870779976715, "step": 4230}, {"loss": 1.4112, "grad_norm": 0.676972508430481, "learning_rate": 0.0002, "epoch": 3.2906480403570044, "step": 4240}, {"loss": 1.4206, "grad_norm": 0.748309314250946, "learning_rate": 0.0002, "epoch": 3.298409002716337, "step": 4250}, {"loss": 1.4973, "grad_norm": 0.6976589560508728, "learning_rate": 0.0002, "epoch": 3.3061699650756693, "step": 4260}, {"loss": 1.3967, "grad_norm": 0.649780809879303, "learning_rate": 0.0002, "epoch": 3.3139309274350017, "step": 4270}, {"loss": 1.327, "grad_norm": 0.6529902815818787, "learning_rate": 0.0002, "epoch": 3.3216918897943346, "step": 4280}, {"loss": 1.4888, "grad_norm": 0.9273163676261902, "learning_rate": 0.0002, "epoch": 3.329452852153667, "step": 4290}, {"loss": 1.4859, "grad_norm": 0.717024028301239, "learning_rate": 0.0002, "epoch": 3.3372138145129995, "step": 4300}, {"loss": 1.4441, "grad_norm": 0.7914950251579285, "learning_rate": 0.0002, "epoch": 3.3449747768723324, "step": 4310}, {"loss": 1.432, "grad_norm": 0.7133203148841858, "learning_rate": 0.0002, "epoch": 3.352735739231665, "step": 4320}, {"loss": 1.4662, "grad_norm": 0.7409568428993225, "learning_rate": 0.0002, "epoch": 3.3604967015909972, "step": 4330}, {"loss": 1.3992, "grad_norm": 0.6993981003761292, "learning_rate": 0.0002, "epoch": 3.3682576639503297, "step": 4340}, {"loss": 1.4261, "grad_norm": 0.7114535570144653, "learning_rate": 0.0002, "epoch": 3.3760186263096625, "step": 4350}, {"loss": 1.4227, "grad_norm": 0.6790860295295715, "learning_rate": 0.0002, "epoch": 3.383779588668995, "step": 4360}, {"loss": 1.4128, "grad_norm": 0.6507849097251892, "learning_rate": 0.0002, "epoch": 3.3915405510283274, "step": 4370}, {"loss": 1.4559, "grad_norm": 0.5967804193496704, "learning_rate": 0.0002, "epoch": 3.39930151338766, "step": 4380}, {"loss": 1.3687, "grad_norm": 0.6625847816467285, "learning_rate": 0.0002, "epoch": 3.4070624757469927, "step": 4390}, {"loss": 1.4193, "grad_norm": 0.6736508011817932, "learning_rate": 0.0002, "epoch": 3.414823438106325, "step": 4400}, {"loss": 1.4363, "grad_norm": 0.7870860695838928, "learning_rate": 0.0002, "epoch": 3.4225844004656576, "step": 4410}, {"loss": 1.4114, "grad_norm": 0.7205295562744141, "learning_rate": 0.0002, "epoch": 3.4303453628249905, "step": 4420}, {"loss": 1.4131, "grad_norm": 0.6634634137153625, "learning_rate": 0.0002, "epoch": 3.438106325184323, "step": 4430}, {"loss": 1.4683, "grad_norm": 0.7562733292579651, "learning_rate": 0.0002, "epoch": 3.4458672875436553, "step": 4440}, {"loss": 1.3486, "grad_norm": 0.6585879921913147, "learning_rate": 0.0002, "epoch": 3.453628249902988, "step": 4450}, {"loss": 1.4283, "grad_norm": 0.6896792054176331, "learning_rate": 0.0002, "epoch": 3.4613892122623207, "step": 4460}, {"loss": 1.4208, "grad_norm": 0.6520342230796814, "learning_rate": 0.0002, "epoch": 3.469150174621653, "step": 4470}, {"loss": 1.3423, "grad_norm": 0.6760806441307068, "learning_rate": 0.0002, "epoch": 3.4769111369809855, "step": 4480}, {"loss": 1.4398, "grad_norm": 0.7539774179458618, "learning_rate": 0.0002, "epoch": 3.484672099340318, "step": 4490}, {"loss": 1.4534, "grad_norm": 0.7409411668777466, "learning_rate": 0.0002, "epoch": 3.492433061699651, "step": 4500}, {"loss": 1.4069, "grad_norm": 0.6876253485679626, "learning_rate": 0.0002, "epoch": 3.5001940240589833, "step": 4510}, {"loss": 1.4228, "grad_norm": 0.7028461694717407, "learning_rate": 0.0002, "epoch": 3.5079549864183157, "step": 4520}, {"loss": 1.4723, "grad_norm": 0.8056529760360718, "learning_rate": 0.0002, "epoch": 3.5157159487776486, "step": 4530}, {"loss": 1.4148, "grad_norm": 0.711338996887207, "learning_rate": 0.0002, "epoch": 3.523476911136981, "step": 4540}, {"loss": 1.5247, "grad_norm": 0.7343552708625793, "learning_rate": 0.0002, "epoch": 3.5312378734963135, "step": 4550}, {"loss": 1.4308, "grad_norm": 0.745479941368103, "learning_rate": 0.0002, "epoch": 3.5389988358556463, "step": 4560}, {"loss": 1.4229, "grad_norm": 0.7582294940948486, "learning_rate": 0.0002, "epoch": 3.5467597982149788, "step": 4570}, {"loss": 1.4127, "grad_norm": 0.6717444658279419, "learning_rate": 0.0002, "epoch": 3.554520760574311, "step": 4580}, {"loss": 1.4368, "grad_norm": 0.7417883276939392, "learning_rate": 0.0002, "epoch": 3.5622817229336436, "step": 4590}, {"loss": 1.4176, "grad_norm": 0.6385737061500549, "learning_rate": 0.0002, "epoch": 3.570042685292976, "step": 4600}, {"loss": 1.3981, "grad_norm": 0.716704249382019, "learning_rate": 0.0002, "epoch": 3.577803647652309, "step": 4610}, {"loss": 1.3889, "grad_norm": 0.6948980093002319, "learning_rate": 0.0002, "epoch": 3.5855646100116414, "step": 4620}, {"loss": 1.5177, "grad_norm": 0.6961140036582947, "learning_rate": 0.0002, "epoch": 3.593325572370974, "step": 4630}, {"loss": 1.4508, "grad_norm": 0.7493122220039368, "learning_rate": 0.0002, "epoch": 3.6010865347303067, "step": 4640}, {"loss": 1.3987, "grad_norm": 0.7431658506393433, "learning_rate": 0.0002, "epoch": 3.608847497089639, "step": 4650}, {"loss": 1.4551, "grad_norm": 0.8353387713432312, "learning_rate": 0.0002, "epoch": 3.6166084594489716, "step": 4660}, {"loss": 1.4533, "grad_norm": 0.7095612287521362, "learning_rate": 0.0002, "epoch": 3.6243694218083045, "step": 4670}, {"loss": 1.4003, "grad_norm": 0.776620090007782, "learning_rate": 0.0002, "epoch": 3.632130384167637, "step": 4680}, {"loss": 1.4361, "grad_norm": 0.7198925018310547, "learning_rate": 0.0002, "epoch": 3.6398913465269693, "step": 4690}, {"loss": 1.4543, "grad_norm": 0.8238834738731384, "learning_rate": 0.0002, "epoch": 3.6476523088863018, "step": 4700}, {"loss": 1.3958, "grad_norm": 0.6804245710372925, "learning_rate": 0.0002, "epoch": 3.655413271245634, "step": 4710}, {"loss": 1.4158, "grad_norm": 0.8444845676422119, "learning_rate": 0.0002, "epoch": 3.663174233604967, "step": 4720}, {"loss": 1.3825, "grad_norm": 0.743797779083252, "learning_rate": 0.0002, "epoch": 3.6709351959642995, "step": 4730}, {"loss": 1.4213, "grad_norm": 0.8994188904762268, "learning_rate": 0.0002, "epoch": 3.678696158323632, "step": 4740}, {"loss": 1.4281, "grad_norm": 0.75416100025177, "learning_rate": 0.0002, "epoch": 3.686457120682965, "step": 4750}, {"loss": 1.4154, "grad_norm": 0.6499266028404236, "learning_rate": 0.0002, "epoch": 3.6942180830422973, "step": 4760}, {"loss": 1.4005, "grad_norm": 0.7246791124343872, "learning_rate": 0.0002, "epoch": 3.7019790454016297, "step": 4770}, {"loss": 1.426, "grad_norm": 0.7831124067306519, "learning_rate": 0.0002, "epoch": 3.7097400077609626, "step": 4780}, {"loss": 1.3933, "grad_norm": 0.7130028009414673, "learning_rate": 0.0002, "epoch": 3.717500970120295, "step": 4790}, {"loss": 1.4632, "grad_norm": 0.7501602172851562, "learning_rate": 0.0002, "epoch": 3.7252619324796274, "step": 4800}, {"loss": 1.4985, "grad_norm": 0.6980932950973511, "learning_rate": 0.0002, "epoch": 3.73302289483896, "step": 4810}, {"loss": 1.4517, "grad_norm": 0.8050530552864075, "learning_rate": 0.0002, "epoch": 3.7407838571982923, "step": 4820}, {"loss": 1.4703, "grad_norm": 0.6385579705238342, "learning_rate": 0.0002, "epoch": 3.748544819557625, "step": 4830}, {"loss": 1.5281, "grad_norm": 0.6664714813232422, "learning_rate": 0.0002, "epoch": 3.7563057819169576, "step": 4840}, {"loss": 1.4443, "grad_norm": 0.7125676274299622, "learning_rate": 0.0002, "epoch": 3.76406674427629, "step": 4850}, {"loss": 1.3958, "grad_norm": 0.7231866717338562, "learning_rate": 0.0002, "epoch": 3.771827706635623, "step": 4860}, {"loss": 1.4446, "grad_norm": 0.6917183995246887, "learning_rate": 0.0002, "epoch": 3.7795886689949554, "step": 4870}, {"loss": 1.4369, "grad_norm": 0.665037989616394, "learning_rate": 0.0002, "epoch": 3.787349631354288, "step": 4880}, {"loss": 1.4193, "grad_norm": 0.5837726593017578, "learning_rate": 0.0002, "epoch": 3.7951105937136207, "step": 4890}, {"loss": 1.4176, "grad_norm": 0.6366701722145081, "learning_rate": 0.0002, "epoch": 3.802871556072953, "step": 4900}, {"loss": 1.46, "grad_norm": 0.7082223892211914, "learning_rate": 0.0002, "epoch": 3.8106325184322856, "step": 4910}, {"loss": 1.5139, "grad_norm": 0.8101672530174255, "learning_rate": 0.0002, "epoch": 3.818393480791618, "step": 4920}, {"loss": 1.3659, "grad_norm": 0.7516148090362549, "learning_rate": 0.0002, "epoch": 3.826154443150951, "step": 4930}, {"loss": 1.3909, "grad_norm": 0.7928489446640015, "learning_rate": 0.0002, "epoch": 3.8339154055102833, "step": 4940}, {"loss": 1.4255, "grad_norm": 0.6892234683036804, "learning_rate": 0.0002, "epoch": 3.8416763678696157, "step": 4950}, {"loss": 1.5024, "grad_norm": 0.6381304264068604, "learning_rate": 0.0002, "epoch": 3.849437330228948, "step": 4960}, {"loss": 1.4873, "grad_norm": 0.8068831562995911, "learning_rate": 0.0002, "epoch": 3.857198292588281, "step": 4970}, {"loss": 1.45, "grad_norm": 0.7289869785308838, "learning_rate": 0.0002, "epoch": 3.8649592549476135, "step": 4980}, {"loss": 1.398, "grad_norm": 0.7278549075126648, "learning_rate": 0.0002, "epoch": 3.872720217306946, "step": 4990}, {"loss": 1.4442, "grad_norm": 0.7324236631393433, "learning_rate": 0.0002, "epoch": 3.880481179666279, "step": 5000}, {"loss": 1.4511, "grad_norm": 0.6759871244430542, "learning_rate": 0.0002, "epoch": 3.8882421420256112, "step": 5010}, {"loss": 1.4705, "grad_norm": 0.8159207701683044, "learning_rate": 0.0002, "epoch": 3.8960031043849437, "step": 5020}, {"loss": 1.4685, "grad_norm": 0.6536211967468262, "learning_rate": 0.0002, "epoch": 3.9037640667442766, "step": 5030}, {"loss": 1.4335, "grad_norm": 0.6827932000160217, "learning_rate": 0.0002, "epoch": 3.911525029103609, "step": 5040}, {"loss": 1.433, "grad_norm": 0.6688340306282043, "learning_rate": 0.0002, "epoch": 3.9192859914629414, "step": 5050}, {"loss": 1.4099, "grad_norm": 0.6385695934295654, "learning_rate": 0.0002, "epoch": 3.927046953822274, "step": 5060}, {"loss": 1.4767, "grad_norm": 0.6975107192993164, "learning_rate": 0.0002, "epoch": 3.9348079161816063, "step": 5070}, {"loss": 1.4893, "grad_norm": 0.6684112548828125, "learning_rate": 0.0002, "epoch": 3.942568878540939, "step": 5080}, {"loss": 1.4732, "grad_norm": 0.8349628448486328, "learning_rate": 0.0002, "epoch": 3.9503298409002716, "step": 5090}, {"loss": 1.5131, "grad_norm": 0.7146425843238831, "learning_rate": 0.0002, "epoch": 3.958090803259604, "step": 5100}, {"loss": 1.4149, "grad_norm": 0.6555036902427673, "learning_rate": 0.0002, "epoch": 3.965851765618937, "step": 5110}, {"loss": 1.4274, "grad_norm": 0.7037415504455566, "learning_rate": 0.0002, "epoch": 3.9736127279782694, "step": 5120}, {"loss": 1.4292, "grad_norm": 0.7235575914382935, "learning_rate": 0.0002, "epoch": 3.981373690337602, "step": 5130}, {"loss": 1.4455, "grad_norm": 0.7092325687408447, "learning_rate": 0.0002, "epoch": 3.9891346526969347, "step": 5140}, {"loss": 1.4512, "grad_norm": 0.7490319609642029, "learning_rate": 0.0002, "epoch": 3.996895615056267, "step": 5150}, {"eval_loss": 1.9131355285644531, "eval_runtime": 105.5778, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 4.0, "step": 5154}, {"loss": 1.2643, "grad_norm": 0.7075854539871216, "learning_rate": 0.0002, "epoch": 4.0046565774155995, "step": 5160}, {"loss": 1.209, "grad_norm": 0.9466007351875305, "learning_rate": 0.0002, "epoch": 4.012417539774932, "step": 5170}, {"loss": 1.2567, "grad_norm": 1.0297044515609741, "learning_rate": 0.0002, "epoch": 4.020178502134264, "step": 5180}, {"loss": 1.1796, "grad_norm": 0.7765059471130371, "learning_rate": 0.0002, "epoch": 4.027939464493597, "step": 5190}, {"loss": 1.2356, "grad_norm": 0.995760977268219, "learning_rate": 0.0002, "epoch": 4.03570042685293, "step": 5200}, {"loss": 1.1792, "grad_norm": 0.8663829565048218, "learning_rate": 0.0002, "epoch": 4.043461389212262, "step": 5210}, {"loss": 1.2471, "grad_norm": 1.0660825967788696, "learning_rate": 0.0002, "epoch": 4.051222351571595, "step": 5220}, {"loss": 1.1676, "grad_norm": 0.9858174920082092, "learning_rate": 0.0002, "epoch": 4.058983313930927, "step": 5230}, {"loss": 1.2448, "grad_norm": 0.8911338448524475, "learning_rate": 0.0002, "epoch": 4.06674427629026, "step": 5240}, {"loss": 1.1858, "grad_norm": 1.0848394632339478, "learning_rate": 0.0002, "epoch": 4.074505238649593, "step": 5250}, {"loss": 1.1684, "grad_norm": 1.0849905014038086, "learning_rate": 0.0002, "epoch": 4.082266201008925, "step": 5260}, {"loss": 1.2007, "grad_norm": 1.0497841835021973, "learning_rate": 0.0002, "epoch": 4.090027163368258, "step": 5270}, {"loss": 1.2552, "grad_norm": 0.8943053483963013, "learning_rate": 0.0002, "epoch": 4.0977881257275905, "step": 5280}, {"loss": 1.1923, "grad_norm": 0.8432527184486389, "learning_rate": 0.0002, "epoch": 4.1055490880869225, "step": 5290}, {"loss": 1.1634, "grad_norm": 0.9690414667129517, "learning_rate": 0.0002, "epoch": 4.113310050446255, "step": 5300}, {"loss": 1.3019, "grad_norm": 0.7790773510932922, "learning_rate": 0.0002, "epoch": 4.121071012805588, "step": 5310}, {"loss": 1.1806, "grad_norm": 0.9289211630821228, "learning_rate": 0.0002, "epoch": 4.12883197516492, "step": 5320}, {"loss": 1.1458, "grad_norm": 1.0785125494003296, "learning_rate": 0.0002, "epoch": 4.136592937524253, "step": 5330}, {"loss": 1.2086, "grad_norm": 0.8559591770172119, "learning_rate": 0.0002, "epoch": 4.144353899883585, "step": 5340}, {"loss": 1.1974, "grad_norm": 0.9405956268310547, "learning_rate": 0.0002, "epoch": 4.152114862242918, "step": 5350}, {"loss": 1.1793, "grad_norm": 0.9942827820777893, "learning_rate": 0.0002, "epoch": 4.159875824602251, "step": 5360}, {"loss": 1.1659, "grad_norm": 0.9141933917999268, "learning_rate": 0.0002, "epoch": 4.167636786961583, "step": 5370}, {"loss": 1.1647, "grad_norm": 0.8206015229225159, "learning_rate": 0.0002, "epoch": 4.175397749320916, "step": 5380}, {"loss": 1.2778, "grad_norm": 0.9340888857841492, "learning_rate": 0.0002, "epoch": 4.183158711680249, "step": 5390}, {"loss": 1.2459, "grad_norm": 1.2122114896774292, "learning_rate": 0.0002, "epoch": 4.190919674039581, "step": 5400}, {"loss": 1.2371, "grad_norm": 1.0661298036575317, "learning_rate": 0.0002, "epoch": 4.1986806363989135, "step": 5410}, {"loss": 1.1978, "grad_norm": 0.9372861385345459, "learning_rate": 0.0002, "epoch": 4.206441598758246, "step": 5420}, {"loss": 1.2653, "grad_norm": 0.894012987613678, "learning_rate": 0.0002, "epoch": 4.214202561117578, "step": 5430}, {"loss": 1.387, "grad_norm": 1.0647753477096558, "learning_rate": 0.0002, "epoch": 4.221963523476911, "step": 5440}, {"loss": 1.2231, "grad_norm": 0.989179790019989, "learning_rate": 0.0002, "epoch": 4.229724485836243, "step": 5450}, {"loss": 1.2715, "grad_norm": 1.1601181030273438, "learning_rate": 0.0002, "epoch": 4.237485448195576, "step": 5460}, {"loss": 1.2406, "grad_norm": 0.9395585656166077, "learning_rate": 0.0002, "epoch": 4.245246410554909, "step": 5470}, {"loss": 1.2779, "grad_norm": 0.9527766108512878, "learning_rate": 0.0002, "epoch": 4.253007372914241, "step": 5480}, {"loss": 1.267, "grad_norm": 1.0319520235061646, "learning_rate": 0.0002, "epoch": 4.260768335273574, "step": 5490}, {"loss": 1.2633, "grad_norm": 0.8659824728965759, "learning_rate": 0.0002, "epoch": 4.268529297632907, "step": 5500}, {"loss": 1.1475, "grad_norm": 1.099211573600769, "learning_rate": 0.0002, "epoch": 4.276290259992239, "step": 5510}, {"loss": 1.2508, "grad_norm": 0.9363361597061157, "learning_rate": 0.0002, "epoch": 4.284051222351572, "step": 5520}, {"loss": 1.189, "grad_norm": 0.8437647223472595, "learning_rate": 0.0002, "epoch": 4.2918121847109045, "step": 5530}, {"loss": 1.2212, "grad_norm": 0.9181258678436279, "learning_rate": 0.0002, "epoch": 4.2995731470702365, "step": 5540}, {"loss": 1.2092, "grad_norm": 0.9059357643127441, "learning_rate": 0.0002, "epoch": 4.307334109429569, "step": 5550}, {"loss": 1.2189, "grad_norm": 0.9337241649627686, "learning_rate": 0.0002, "epoch": 4.315095071788901, "step": 5560}, {"loss": 1.2462, "grad_norm": 0.9428889155387878, "learning_rate": 0.0002, "epoch": 4.322856034148234, "step": 5570}, {"loss": 1.2675, "grad_norm": 1.003589153289795, "learning_rate": 0.0002, "epoch": 4.330616996507567, "step": 5580}, {"loss": 1.2703, "grad_norm": 1.1249268054962158, "learning_rate": 0.0002, "epoch": 4.338377958866899, "step": 5590}, {"loss": 1.2501, "grad_norm": 0.8623469471931458, "learning_rate": 0.0002, "epoch": 4.346138921226232, "step": 5600}, {"loss": 1.2404, "grad_norm": 1.1389174461364746, "learning_rate": 0.0002, "epoch": 4.353899883585565, "step": 5610}, {"loss": 1.2245, "grad_norm": 1.0136264562606812, "learning_rate": 0.0002, "epoch": 4.361660845944897, "step": 5620}, {"loss": 1.3473, "grad_norm": 0.9567070603370667, "learning_rate": 0.0002, "epoch": 4.36942180830423, "step": 5630}, {"loss": 1.2988, "grad_norm": 1.0592148303985596, "learning_rate": 0.0002, "epoch": 4.377182770663563, "step": 5640}, {"loss": 1.212, "grad_norm": 1.0110485553741455, "learning_rate": 0.0002, "epoch": 4.384943733022895, "step": 5650}, {"loss": 1.2086, "grad_norm": 0.9914907217025757, "learning_rate": 0.0002, "epoch": 4.3927046953822275, "step": 5660}, {"loss": 1.2363, "grad_norm": 0.9447247982025146, "learning_rate": 0.0002, "epoch": 4.4004656577415595, "step": 5670}, {"loss": 1.2617, "grad_norm": 0.9644378423690796, "learning_rate": 0.0002, "epoch": 4.408226620100892, "step": 5680}, {"loss": 1.2773, "grad_norm": 0.920676589012146, "learning_rate": 0.0002, "epoch": 4.415987582460225, "step": 5690}, {"loss": 1.2792, "grad_norm": 1.060570478439331, "learning_rate": 0.0002, "epoch": 4.423748544819557, "step": 5700}, {"loss": 1.2374, "grad_norm": 0.8857738971710205, "learning_rate": 0.0002, "epoch": 4.43150950717889, "step": 5710}, {"loss": 1.2588, "grad_norm": 1.0536398887634277, "learning_rate": 0.0002, "epoch": 4.439270469538223, "step": 5720}, {"loss": 1.2051, "grad_norm": 0.990847110748291, "learning_rate": 0.0002, "epoch": 4.447031431897555, "step": 5730}, {"loss": 1.2469, "grad_norm": 0.9692499041557312, "learning_rate": 0.0002, "epoch": 4.454792394256888, "step": 5740}, {"loss": 1.2269, "grad_norm": 1.0376402139663696, "learning_rate": 0.0002, "epoch": 4.462553356616221, "step": 5750}, {"loss": 1.1701, "grad_norm": 1.3863259553909302, "learning_rate": 0.0002, "epoch": 4.470314318975553, "step": 5760}, {"loss": 1.2591, "grad_norm": 0.978379487991333, "learning_rate": 0.0002, "epoch": 4.478075281334886, "step": 5770}, {"loss": 1.2729, "grad_norm": 1.0973085165023804, "learning_rate": 0.0002, "epoch": 4.485836243694218, "step": 5780}, {"loss": 1.2404, "grad_norm": 1.057006597518921, "learning_rate": 0.0002, "epoch": 4.4935972060535505, "step": 5790}, {"loss": 1.2476, "grad_norm": 0.9247729182243347, "learning_rate": 0.0002, "epoch": 4.501358168412883, "step": 5800}, {"loss": 1.2369, "grad_norm": 1.0447787046432495, "learning_rate": 0.0002, "epoch": 4.509119130772215, "step": 5810}, {"loss": 1.211, "grad_norm": 1.1930429935455322, "learning_rate": 0.0002, "epoch": 4.516880093131548, "step": 5820}, {"loss": 1.2596, "grad_norm": 0.9867590069770813, "learning_rate": 0.0002, "epoch": 4.524641055490881, "step": 5830}, {"loss": 1.2766, "grad_norm": 0.9591100215911865, "learning_rate": 0.0002, "epoch": 4.532402017850213, "step": 5840}, {"loss": 1.2154, "grad_norm": 0.9950753450393677, "learning_rate": 0.0002, "epoch": 4.540162980209546, "step": 5850}, {"loss": 1.2149, "grad_norm": 1.0087506771087646, "learning_rate": 0.0002, "epoch": 4.547923942568879, "step": 5860}, {"loss": 1.3165, "grad_norm": 1.0934417247772217, "learning_rate": 0.0002, "epoch": 4.555684904928211, "step": 5870}, {"loss": 1.3059, "grad_norm": 1.107987403869629, "learning_rate": 0.0002, "epoch": 4.563445867287544, "step": 5880}, {"loss": 1.2184, "grad_norm": 0.9147276878356934, "learning_rate": 0.0002, "epoch": 4.571206829646876, "step": 5890}, {"loss": 1.24, "grad_norm": 1.036780595779419, "learning_rate": 0.0002, "epoch": 4.578967792006209, "step": 5900}, {"loss": 1.2209, "grad_norm": 0.9284719824790955, "learning_rate": 0.0002, "epoch": 4.5867287543655415, "step": 5910}, {"loss": 1.3693, "grad_norm": 0.9141898155212402, "learning_rate": 0.0002, "epoch": 4.5944897167248735, "step": 5920}, {"loss": 1.2319, "grad_norm": 1.0447357892990112, "learning_rate": 0.0002, "epoch": 4.602250679084206, "step": 5930}, {"loss": 1.2667, "grad_norm": 0.9309114217758179, "learning_rate": 0.0002, "epoch": 4.610011641443539, "step": 5940}, {"loss": 1.2827, "grad_norm": 1.2986129522323608, "learning_rate": 0.0002, "epoch": 4.617772603802871, "step": 5950}, {"loss": 1.312, "grad_norm": 0.9221704602241516, "learning_rate": 0.0002, "epoch": 4.625533566162204, "step": 5960}, {"loss": 1.2769, "grad_norm": 0.9228187799453735, "learning_rate": 0.0002, "epoch": 4.633294528521537, "step": 5970}, {"loss": 1.2953, "grad_norm": 0.9483116269111633, "learning_rate": 0.0002, "epoch": 4.641055490880869, "step": 5980}, {"loss": 1.3437, "grad_norm": 1.0218974351882935, "learning_rate": 0.0002, "epoch": 4.648816453240202, "step": 5990}, {"loss": 1.3085, "grad_norm": 0.9764600396156311, "learning_rate": 0.0002, "epoch": 4.656577415599534, "step": 6000}, {"loss": 1.197, "grad_norm": 0.9115710258483887, "learning_rate": 0.0002, "epoch": 4.664338377958867, "step": 6010}, {"loss": 1.1917, "grad_norm": 0.9245651364326477, "learning_rate": 0.0002, "epoch": 4.6720993403182, "step": 6020}, {"loss": 1.2969, "grad_norm": 0.9686311483383179, "learning_rate": 0.0002, "epoch": 4.6798603026775325, "step": 6030}, {"loss": 1.2702, "grad_norm": 1.1807392835617065, "learning_rate": 0.0002, "epoch": 4.6876212650368645, "step": 6040}, {"loss": 1.328, "grad_norm": 1.0358641147613525, "learning_rate": 0.0002, "epoch": 4.695382227396197, "step": 6050}, {"loss": 1.3281, "grad_norm": 0.987332284450531, "learning_rate": 0.0002, "epoch": 4.703143189755529, "step": 6060}, {"loss": 1.2514, "grad_norm": 1.0526494979858398, "learning_rate": 0.0002, "epoch": 4.710904152114862, "step": 6070}, {"loss": 1.2246, "grad_norm": 1.0276758670806885, "learning_rate": 0.0002, "epoch": 4.718665114474195, "step": 6080}, {"loss": 1.3367, "grad_norm": 0.9904406666755676, "learning_rate": 0.0002, "epoch": 4.726426076833527, "step": 6090}, {"loss": 1.2797, "grad_norm": 1.0084882974624634, "learning_rate": 0.0002, "epoch": 4.73418703919286, "step": 6100}, {"loss": 1.2656, "grad_norm": 0.8646450638771057, "learning_rate": 0.0002, "epoch": 4.741948001552192, "step": 6110}, {"loss": 1.3063, "grad_norm": 0.9233377575874329, "learning_rate": 0.0002, "epoch": 4.749708963911525, "step": 6120}, {"loss": 1.2642, "grad_norm": 0.9675140976905823, "learning_rate": 0.0002, "epoch": 4.757469926270858, "step": 6130}, {"loss": 1.3367, "grad_norm": 0.9639796018600464, "learning_rate": 0.0002, "epoch": 4.765230888630191, "step": 6140}, {"loss": 1.276, "grad_norm": 0.925199568271637, "learning_rate": 0.0002, "epoch": 4.772991850989523, "step": 6150}, {"loss": 1.2441, "grad_norm": 1.050901174545288, "learning_rate": 0.0002, "epoch": 4.7807528133488555, "step": 6160}, {"loss": 1.301, "grad_norm": 0.8920623660087585, "learning_rate": 0.0002, "epoch": 4.7885137757081875, "step": 6170}, {"loss": 1.263, "grad_norm": 0.8964757919311523, "learning_rate": 0.0002, "epoch": 4.79627473806752, "step": 6180}, {"loss": 1.2787, "grad_norm": 1.0839070081710815, "learning_rate": 0.0002, "epoch": 4.804035700426853, "step": 6190}, {"loss": 1.2664, "grad_norm": 0.8809942007064819, "learning_rate": 0.0002, "epoch": 4.811796662786185, "step": 6200}, {"loss": 1.321, "grad_norm": 1.0216195583343506, "learning_rate": 0.0002, "epoch": 4.819557625145518, "step": 6210}, {"loss": 1.3033, "grad_norm": 0.892005980014801, "learning_rate": 0.0002, "epoch": 4.827318587504851, "step": 6220}, {"loss": 1.2602, "grad_norm": 0.9957166910171509, "learning_rate": 0.0002, "epoch": 4.835079549864183, "step": 6230}, {"loss": 1.3562, "grad_norm": 0.9720533490180969, "learning_rate": 0.0002, "epoch": 4.842840512223516, "step": 6240}, {"loss": 1.2651, "grad_norm": 0.9336182475090027, "learning_rate": 0.0002, "epoch": 4.850601474582849, "step": 6250}, {"loss": 1.3136, "grad_norm": 1.2611457109451294, "learning_rate": 0.0002, "epoch": 4.858362436942181, "step": 6260}, {"loss": 1.2234, "grad_norm": 0.8927203416824341, "learning_rate": 0.0002, "epoch": 4.866123399301514, "step": 6270}, {"loss": 1.3463, "grad_norm": 0.9706710577011108, "learning_rate": 0.0002, "epoch": 4.873884361660846, "step": 6280}, {"loss": 1.3209, "grad_norm": 1.1461690664291382, "learning_rate": 0.0002, "epoch": 4.8816453240201785, "step": 6290}, {"loss": 1.2566, "grad_norm": 0.9930381178855896, "learning_rate": 0.0002, "epoch": 4.889406286379511, "step": 6300}, {"loss": 1.2568, "grad_norm": 0.91451096534729, "learning_rate": 0.0002, "epoch": 4.897167248738843, "step": 6310}, {"loss": 1.2836, "grad_norm": 1.0319571495056152, "learning_rate": 0.0002, "epoch": 4.904928211098176, "step": 6320}, {"loss": 1.2908, "grad_norm": 0.990140438079834, "learning_rate": 0.0002, "epoch": 4.912689173457509, "step": 6330}, {"loss": 1.3299, "grad_norm": 1.2466117143630981, "learning_rate": 0.0002, "epoch": 4.920450135816841, "step": 6340}, {"loss": 1.2659, "grad_norm": 1.0316979885101318, "learning_rate": 0.0002, "epoch": 4.928211098176174, "step": 6350}, {"loss": 1.3292, "grad_norm": 1.0643759965896606, "learning_rate": 0.0002, "epoch": 4.935972060535507, "step": 6360}, {"loss": 1.2559, "grad_norm": 0.9703279733657837, "learning_rate": 0.0002, "epoch": 4.943733022894839, "step": 6370}, {"loss": 1.2155, "grad_norm": 0.9767927527427673, "learning_rate": 0.0002, "epoch": 4.951493985254172, "step": 6380}, {"loss": 1.2437, "grad_norm": 0.960854172706604, "learning_rate": 0.0002, "epoch": 4.959254947613504, "step": 6390}, {"loss": 1.3314, "grad_norm": 0.9922910332679749, "learning_rate": 0.0002, "epoch": 4.967015909972837, "step": 6400}, {"loss": 1.3018, "grad_norm": 0.956470787525177, "learning_rate": 0.0002, "epoch": 4.9747768723321695, "step": 6410}, {"loss": 1.2794, "grad_norm": 0.9637242555618286, "learning_rate": 0.0002, "epoch": 4.9825378346915015, "step": 6420}, {"loss": 1.3236, "grad_norm": 1.0855202674865723, "learning_rate": 0.0002, "epoch": 4.990298797050834, "step": 6430}, {"loss": 1.3015, "grad_norm": 0.9655316472053528, "learning_rate": 0.0002, "epoch": 4.998059759410167, "step": 6440}, {"eval_loss": 2.0410802364349365, "eval_runtime": 113.04, "eval_samples_per_second": 4.485, "eval_steps_per_second": 0.566, "epoch": 4.9996119518820334, "step": 6442}, {"loss": 1.0846, "grad_norm": 1.1676199436187744, "learning_rate": 0.0002, "epoch": 5.005820721769499, "step": 6450}, {"loss": 1.041, "grad_norm": 1.4317965507507324, "learning_rate": 0.0002, "epoch": 5.013581684128832, "step": 6460}, {"loss": 0.9546, "grad_norm": 1.460443377494812, "learning_rate": 0.0002, "epoch": 5.021342646488165, "step": 6470}, {"loss": 1.0014, "grad_norm": 1.2299214601516724, "learning_rate": 0.0002, "epoch": 5.029103608847497, "step": 6480}, {"loss": 1.0397, "grad_norm": 1.3125724792480469, "learning_rate": 0.0002, "epoch": 5.03686457120683, "step": 6490}, {"loss": 1.0134, "grad_norm": 1.1252319812774658, "learning_rate": 0.0002, "epoch": 5.044625533566162, "step": 6500}, {"loss": 0.976, "grad_norm": 0.9970866441726685, "learning_rate": 0.0002, "epoch": 5.052386495925495, "step": 6510}, {"loss": 0.9731, "grad_norm": 1.229069709777832, "learning_rate": 0.0002, "epoch": 5.060147458284828, "step": 6520}, {"loss": 1.0498, "grad_norm": 1.2430938482284546, "learning_rate": 0.0002, "epoch": 5.06790842064416, "step": 6530}, {"loss": 1.0236, "grad_norm": 1.0522737503051758, "learning_rate": 0.0002, "epoch": 5.0756693830034925, "step": 6540}, {"loss": 1.0221, "grad_norm": 1.108890175819397, "learning_rate": 0.0002, "epoch": 5.083430345362825, "step": 6550}, {"loss": 1.0177, "grad_norm": 1.156912922859192, "learning_rate": 0.0002, "epoch": 5.091191307722157, "step": 6560}, {"loss": 1.0415, "grad_norm": 1.405895709991455, "learning_rate": 0.0002, "epoch": 5.09895227008149, "step": 6570}, {"loss": 0.9811, "grad_norm": 1.2005155086517334, "learning_rate": 0.0002, "epoch": 5.106713232440823, "step": 6580}, {"loss": 0.9862, "grad_norm": 1.181443452835083, "learning_rate": 0.0002, "epoch": 5.114474194800155, "step": 6590}, {"loss": 1.0291, "grad_norm": 2.3444771766662598, "learning_rate": 0.0002, "epoch": 5.122235157159488, "step": 6600}, {"loss": 1.0455, "grad_norm": 1.216988444328308, "learning_rate": 0.0002, "epoch": 5.12999611951882, "step": 6610}, {"loss": 1.0549, "grad_norm": 1.369553565979004, "learning_rate": 0.0002, "epoch": 5.137757081878153, "step": 6620}, {"loss": 1.0056, "grad_norm": 1.177964687347412, "learning_rate": 0.0002, "epoch": 5.145518044237486, "step": 6630}, {"loss": 1.1025, "grad_norm": 1.1397041082382202, "learning_rate": 0.0002, "epoch": 5.153279006596818, "step": 6640}, {"loss": 1.0437, "grad_norm": 1.3976861238479614, "learning_rate": 0.0002, "epoch": 5.161039968956151, "step": 6650}, {"loss": 1.0454, "grad_norm": 1.4824495315551758, "learning_rate": 0.0002, "epoch": 5.1688009313154835, "step": 6660}, {"loss": 1.0356, "grad_norm": 1.2653018236160278, "learning_rate": 0.0002, "epoch": 5.1765618936748155, "step": 6670}, {"loss": 0.9971, "grad_norm": 1.3106069564819336, "learning_rate": 0.0002, "epoch": 5.184322856034148, "step": 6680}, {"loss": 1.0561, "grad_norm": 1.3140279054641724, "learning_rate": 0.0002, "epoch": 5.192083818393481, "step": 6690}, {"loss": 1.0618, "grad_norm": 1.3900256156921387, "learning_rate": 0.0002, "epoch": 5.199844780752813, "step": 6700}, {"loss": 1.0285, "grad_norm": 1.3191124200820923, "learning_rate": 0.0002, "epoch": 5.207605743112146, "step": 6710}, {"loss": 0.9921, "grad_norm": 1.176107406616211, "learning_rate": 0.0002, "epoch": 5.215366705471478, "step": 6720}, {"loss": 1.064, "grad_norm": 1.2364883422851562, "learning_rate": 0.0002, "epoch": 5.223127667830811, "step": 6730}, {"loss": 0.9599, "grad_norm": 1.343022108078003, "learning_rate": 0.0002, "epoch": 5.230888630190144, "step": 6740}, {"loss": 1.0342, "grad_norm": 1.2826898097991943, "learning_rate": 0.0002, "epoch": 5.238649592549476, "step": 6750}, {"loss": 1.0703, "grad_norm": 1.500257134437561, "learning_rate": 0.0002, "epoch": 5.246410554908809, "step": 6760}, {"loss": 1.0114, "grad_norm": 1.2605743408203125, "learning_rate": 0.0002, "epoch": 5.254171517268142, "step": 6770}, {"loss": 1.0825, "grad_norm": 1.2355525493621826, "learning_rate": 0.0002, "epoch": 5.261932479627474, "step": 6780}, {"loss": 1.0436, "grad_norm": 1.2845789194107056, "learning_rate": 0.0002, "epoch": 5.2696934419868064, "step": 6790}, {"loss": 0.989, "grad_norm": 1.3696625232696533, "learning_rate": 0.0002, "epoch": 5.277454404346139, "step": 6800}, {"loss": 1.0991, "grad_norm": 1.4051260948181152, "learning_rate": 0.0002, "epoch": 5.285215366705471, "step": 6810}, {"loss": 1.0987, "grad_norm": 1.266725778579712, "learning_rate": 0.0002, "epoch": 5.292976329064804, "step": 6820}, {"loss": 1.0489, "grad_norm": 1.3475236892700195, "learning_rate": 0.0002, "epoch": 5.300737291424136, "step": 6830}, {"loss": 1.0264, "grad_norm": 1.54409921169281, "learning_rate": 0.0002, "epoch": 5.308498253783469, "step": 6840}, {"loss": 1.033, "grad_norm": 1.2391985654830933, "learning_rate": 0.0002, "epoch": 5.316259216142802, "step": 6850}, {"loss": 1.1058, "grad_norm": 1.2435699701309204, "learning_rate": 0.0002, "epoch": 5.324020178502134, "step": 6860}, {"loss": 1.0179, "grad_norm": 1.8803037405014038, "learning_rate": 0.0002, "epoch": 5.331781140861467, "step": 6870}, {"loss": 0.997, "grad_norm": 1.4195542335510254, "learning_rate": 0.0002, "epoch": 5.3395421032208, "step": 6880}, {"loss": 1.0273, "grad_norm": 1.1853394508361816, "learning_rate": 0.0002, "epoch": 5.347303065580132, "step": 6890}, {"loss": 1.0668, "grad_norm": 1.4016530513763428, "learning_rate": 0.0002, "epoch": 5.355064027939465, "step": 6900}, {"loss": 1.1099, "grad_norm": 1.294339895248413, "learning_rate": 0.0002, "epoch": 5.3628249902987974, "step": 6910}, {"loss": 1.0724, "grad_norm": 1.2952708005905151, "learning_rate": 0.0002, "epoch": 5.370585952658129, "step": 6920}, {"loss": 1.0098, "grad_norm": 1.1361510753631592, "learning_rate": 0.0002, "epoch": 5.378346915017462, "step": 6930}, {"loss": 1.0796, "grad_norm": 1.125805377960205, "learning_rate": 0.0002, "epoch": 5.386107877376794, "step": 6940}, {"loss": 1.122, "grad_norm": 1.1453300714492798, "learning_rate": 0.0002, "epoch": 5.393868839736127, "step": 6950}, {"loss": 1.0977, "grad_norm": 1.4542768001556396, "learning_rate": 0.0002, "epoch": 5.40162980209546, "step": 6960}, {"loss": 1.0825, "grad_norm": 1.2360988855361938, "learning_rate": 0.0002, "epoch": 5.409390764454792, "step": 6970}, {"loss": 1.0631, "grad_norm": 1.2182754278182983, "learning_rate": 0.0002, "epoch": 5.417151726814125, "step": 6980}, {"loss": 1.0471, "grad_norm": 1.2018693685531616, "learning_rate": 0.0002, "epoch": 5.424912689173458, "step": 6990}, {"loss": 1.108, "grad_norm": 1.346124291419983, "learning_rate": 0.0002, "epoch": 5.43267365153279, "step": 7000}, {"loss": 1.0534, "grad_norm": 1.2534189224243164, "learning_rate": 0.0002, "epoch": 5.440434613892123, "step": 7010}, {"loss": 1.0696, "grad_norm": 1.2033339738845825, "learning_rate": 0.0002, "epoch": 5.448195576251456, "step": 7020}, {"loss": 1.0714, "grad_norm": 1.2788134813308716, "learning_rate": 0.0002, "epoch": 5.4559565386107876, "step": 7030}, {"loss": 1.1274, "grad_norm": 1.2751542329788208, "learning_rate": 0.0002, "epoch": 5.46371750097012, "step": 7040}, {"loss": 1.0767, "grad_norm": 1.3237019777297974, "learning_rate": 0.0002, "epoch": 5.471478463329452, "step": 7050}, {"loss": 1.1081, "grad_norm": 1.4932852983474731, "learning_rate": 0.0002, "epoch": 5.479239425688785, "step": 7060}, {"loss": 1.0197, "grad_norm": 1.4003876447677612, "learning_rate": 0.0002, "epoch": 5.487000388048118, "step": 7070}, {"loss": 1.0662, "grad_norm": 1.404799461364746, "learning_rate": 0.0002, "epoch": 5.49476135040745, "step": 7080}, {"loss": 1.0354, "grad_norm": 1.4486982822418213, "learning_rate": 0.0002, "epoch": 5.502522312766783, "step": 7090}, {"loss": 1.0645, "grad_norm": 1.1713480949401855, "learning_rate": 0.0002, "epoch": 5.510283275126116, "step": 7100}, {"loss": 1.006, "grad_norm": 1.4062601327896118, "learning_rate": 0.0002, "epoch": 5.518044237485448, "step": 7110}, {"loss": 1.0459, "grad_norm": 1.211629867553711, "learning_rate": 0.0002, "epoch": 5.525805199844781, "step": 7120}, {"loss": 1.102, "grad_norm": 1.2523176670074463, "learning_rate": 0.0002, "epoch": 5.533566162204114, "step": 7130}, {"loss": 1.1132, "grad_norm": 1.4467198848724365, "learning_rate": 0.0002, "epoch": 5.541327124563446, "step": 7140}, {"loss": 1.1557, "grad_norm": 1.5961614847183228, "learning_rate": 0.0002, "epoch": 5.5490880869227786, "step": 7150}, {"loss": 1.0859, "grad_norm": 1.320656418800354, "learning_rate": 0.0002, "epoch": 5.5568490492821105, "step": 7160}, {"loss": 1.109, "grad_norm": 1.2423332929611206, "learning_rate": 0.0002, "epoch": 5.564610011641443, "step": 7170}, {"loss": 1.0046, "grad_norm": 1.2919669151306152, "learning_rate": 0.0002, "epoch": 5.572370974000776, "step": 7180}, {"loss": 1.046, "grad_norm": 1.1678385734558105, "learning_rate": 0.0002, "epoch": 5.580131936360108, "step": 7190}, {"loss": 1.1011, "grad_norm": 1.4250764846801758, "learning_rate": 0.0002, "epoch": 5.587892898719441, "step": 7200}, {"loss": 1.1254, "grad_norm": 1.5308716297149658, "learning_rate": 0.0002, "epoch": 5.595653861078774, "step": 7210}, {"loss": 1.121, "grad_norm": 1.2678815126419067, "learning_rate": 0.0002, "epoch": 5.603414823438106, "step": 7220}, {"loss": 1.0846, "grad_norm": 1.127856969833374, "learning_rate": 0.0002, "epoch": 5.611175785797439, "step": 7230}, {"loss": 1.0647, "grad_norm": 1.3832560777664185, "learning_rate": 0.0002, "epoch": 5.618936748156772, "step": 7240}, {"loss": 1.0658, "grad_norm": 1.3226919174194336, "learning_rate": 0.0002, "epoch": 5.626697710516104, "step": 7250}, {"loss": 1.1175, "grad_norm": 1.3418006896972656, "learning_rate": 0.0002, "epoch": 5.634458672875437, "step": 7260}, {"loss": 1.0956, "grad_norm": 1.2625300884246826, "learning_rate": 0.0002, "epoch": 5.642219635234769, "step": 7270}, {"loss": 1.067, "grad_norm": 1.1579464673995972, "learning_rate": 0.0002, "epoch": 5.6499805975941015, "step": 7280}, {"loss": 1.0447, "grad_norm": 1.4998650550842285, "learning_rate": 0.0002, "epoch": 5.657741559953434, "step": 7290}, {"loss": 1.1256, "grad_norm": 1.2670758962631226, "learning_rate": 0.0002, "epoch": 5.665502522312766, "step": 7300}, {"loss": 1.1267, "grad_norm": 1.2959760427474976, "learning_rate": 0.0002, "epoch": 5.673263484672099, "step": 7310}, {"loss": 1.1387, "grad_norm": 1.2460671663284302, "learning_rate": 0.0002, "epoch": 5.681024447031432, "step": 7320}, {"loss": 1.0756, "grad_norm": 1.1313989162445068, "learning_rate": 0.0002, "epoch": 5.688785409390764, "step": 7330}, {"loss": 1.0618, "grad_norm": 1.282527208328247, "learning_rate": 0.0002, "epoch": 5.696546371750097, "step": 7340}, {"loss": 1.1315, "grad_norm": 1.3380206823349, "learning_rate": 0.0002, "epoch": 5.70430733410943, "step": 7350}, {"loss": 1.0949, "grad_norm": 1.1648279428482056, "learning_rate": 0.0002, "epoch": 5.712068296468762, "step": 7360}, {"loss": 1.1705, "grad_norm": 1.3059816360473633, "learning_rate": 0.0002, "epoch": 5.719829258828095, "step": 7370}, {"loss": 1.1496, "grad_norm": 1.1905046701431274, "learning_rate": 0.0002, "epoch": 5.727590221187427, "step": 7380}, {"loss": 1.1356, "grad_norm": 1.4089630842208862, "learning_rate": 0.0002, "epoch": 5.73535118354676, "step": 7390}, {"loss": 1.1349, "grad_norm": 1.256721019744873, "learning_rate": 0.0002, "epoch": 5.7431121459060925, "step": 7400}, {"loss": 1.0682, "grad_norm": 1.1915162801742554, "learning_rate": 0.0002, "epoch": 5.7508731082654245, "step": 7410}, {"loss": 1.1257, "grad_norm": 1.1935480833053589, "learning_rate": 0.0002, "epoch": 5.758634070624757, "step": 7420}, {"loss": 1.1348, "grad_norm": 1.1761008501052856, "learning_rate": 0.0002, "epoch": 5.76639503298409, "step": 7430}, {"loss": 1.0837, "grad_norm": 1.2540549039840698, "learning_rate": 0.0002, "epoch": 5.774155995343422, "step": 7440}, {"loss": 1.1527, "grad_norm": 1.5295120477676392, "learning_rate": 0.0002, "epoch": 5.781916957702755, "step": 7450}, {"loss": 1.1146, "grad_norm": 1.1081160306930542, "learning_rate": 0.0002, "epoch": 5.789677920062088, "step": 7460}, {"loss": 1.1304, "grad_norm": 1.4381253719329834, "learning_rate": 0.0002, "epoch": 5.79743888242142, "step": 7470}, {"loss": 1.0684, "grad_norm": 1.3079341650009155, "learning_rate": 0.0002, "epoch": 5.805199844780753, "step": 7480}, {"loss": 1.0544, "grad_norm": 1.1372792720794678, "learning_rate": 0.0002, "epoch": 5.812960807140085, "step": 7490}, {"loss": 1.1622, "grad_norm": 1.3221744298934937, "learning_rate": 0.0002, "epoch": 5.820721769499418, "step": 7500}, {"loss": 1.1515, "grad_norm": 1.3436939716339111, "learning_rate": 0.0002, "epoch": 5.828482731858751, "step": 7510}, {"loss": 1.1154, "grad_norm": 1.3916879892349243, "learning_rate": 0.0002, "epoch": 5.8362436942180835, "step": 7520}, {"loss": 1.0816, "grad_norm": 1.2463704347610474, "learning_rate": 0.0002, "epoch": 5.8440046565774155, "step": 7530}, {"loss": 1.0745, "grad_norm": 1.097051739692688, "learning_rate": 0.0002, "epoch": 5.851765618936748, "step": 7540}, {"loss": 1.1454, "grad_norm": 1.1554739475250244, "learning_rate": 0.0002, "epoch": 5.85952658129608, "step": 7550}, {"loss": 1.0953, "grad_norm": 1.2384694814682007, "learning_rate": 0.0002, "epoch": 5.867287543655413, "step": 7560}, {"loss": 1.1734, "grad_norm": 1.142815351486206, "learning_rate": 0.0002, "epoch": 5.875048506014746, "step": 7570}, {"loss": 1.162, "grad_norm": 1.3637062311172485, "learning_rate": 0.0002, "epoch": 5.882809468374078, "step": 7580}, {"loss": 1.0781, "grad_norm": 1.2449073791503906, "learning_rate": 0.0002, "epoch": 5.890570430733411, "step": 7590}, {"loss": 1.1191, "grad_norm": 1.358058214187622, "learning_rate": 0.0002, "epoch": 5.898331393092743, "step": 7600}, {"loss": 1.0779, "grad_norm": 1.264655351638794, "learning_rate": 0.0002, "epoch": 5.906092355452076, "step": 7610}, {"loss": 1.1538, "grad_norm": 1.3186019659042358, "learning_rate": 0.0002, "epoch": 5.913853317811409, "step": 7620}, {"loss": 1.1076, "grad_norm": 1.4111460447311401, "learning_rate": 0.0002, "epoch": 5.921614280170742, "step": 7630}, {"loss": 1.1765, "grad_norm": 1.1078972816467285, "learning_rate": 0.0002, "epoch": 5.929375242530074, "step": 7640}, {"loss": 1.1305, "grad_norm": 1.2742213010787964, "learning_rate": 0.0002, "epoch": 5.9371362048894065, "step": 7650}, {"loss": 1.144, "grad_norm": 1.3412781953811646, "learning_rate": 0.0002, "epoch": 5.9448971672487385, "step": 7660}, {"loss": 1.1642, "grad_norm": 1.123005986213684, "learning_rate": 0.0002, "epoch": 5.952658129608071, "step": 7670}, {"loss": 1.0732, "grad_norm": 1.2203444242477417, "learning_rate": 0.0002, "epoch": 5.960419091967404, "step": 7680}, {"loss": 1.158, "grad_norm": 1.341011643409729, "learning_rate": 0.0002, "epoch": 5.968180054326736, "step": 7690}, {"loss": 1.1144, "grad_norm": 1.2689454555511475, "learning_rate": 0.0002, "epoch": 5.975941016686069, "step": 7700}, {"loss": 1.2051, "grad_norm": 1.1518112421035767, "learning_rate": 0.0002, "epoch": 5.983701979045401, "step": 7710}, {"loss": 1.1868, "grad_norm": 1.3698320388793945, "learning_rate": 0.0002, "epoch": 5.991462941404734, "step": 7720}, {"loss": 1.0651, "grad_norm": 1.2812788486480713, "learning_rate": 0.0002, "epoch": 5.999223903764067, "step": 7730}]}
+{"epoch": 6.9996119518820334, "step": 9019, "epoch_duration": 3941.118304014206, "total_accumulated_duration": 25912.183322429657, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}, {"eval_loss": 1.8472607135772705, "eval_runtime": 106.5541, "eval_samples_per_second": 4.758, "eval_steps_per_second": 0.601, "epoch": 2.9996119518820334, "step": 3865}, {"loss": 1.4983, "grad_norm": 0.5645653605461121, "learning_rate": 0.0002, "epoch": 3.0034924330616994, "step": 3870}, {"loss": 1.4334, "grad_norm": 0.6457151174545288, "learning_rate": 0.0002, "epoch": 3.0112533954210323, "step": 3880}, {"loss": 1.3899, "grad_norm": 0.583838164806366, "learning_rate": 0.0002, "epoch": 3.0190143577803648, "step": 3890}, {"loss": 1.3258, "grad_norm": 0.6819260120391846, "learning_rate": 0.0002, "epoch": 3.026775320139697, "step": 3900}, {"loss": 1.3458, "grad_norm": 0.6692903637886047, "learning_rate": 0.0002, "epoch": 3.03453628249903, "step": 3910}, {"loss": 1.4356, "grad_norm": 0.6101024746894836, "learning_rate": 0.0002, "epoch": 3.0422972448583625, "step": 3920}, {"loss": 1.394, "grad_norm": 0.7014093399047852, "learning_rate": 0.0002, "epoch": 3.050058207217695, "step": 3930}, {"loss": 1.3885, "grad_norm": 0.7380381226539612, "learning_rate": 0.0002, "epoch": 3.0578191695770274, "step": 3940}, {"loss": 1.4206, "grad_norm": 0.6607900857925415, "learning_rate": 0.0002, "epoch": 3.0655801319363603, "step": 3950}, {"loss": 1.4293, "grad_norm": 0.735263466835022, "learning_rate": 0.0002, "epoch": 3.0733410942956927, "step": 3960}, {"loss": 1.3966, "grad_norm": 0.6788513660430908, "learning_rate": 0.0002, "epoch": 3.081102056655025, "step": 3970}, {"loss": 1.3435, "grad_norm": 0.6347652673721313, "learning_rate": 0.0002, "epoch": 3.088863019014358, "step": 3980}, {"loss": 1.4518, "grad_norm": 0.7056642770767212, "learning_rate": 0.0002, "epoch": 3.0966239813736904, "step": 3990}, {"loss": 1.4474, "grad_norm": 0.6387075185775757, "learning_rate": 0.0002, "epoch": 3.104384943733023, "step": 4000}, {"loss": 1.3833, "grad_norm": 0.6701116561889648, "learning_rate": 0.0002, "epoch": 3.1121459060923553, "step": 4010}, {"loss": 1.404, "grad_norm": 0.7558449506759644, "learning_rate": 0.0002, "epoch": 3.119906868451688, "step": 4020}, {"loss": 1.3294, "grad_norm": 0.6612881422042847, "learning_rate": 0.0002, "epoch": 3.1276678308110206, "step": 4030}, {"loss": 1.439, "grad_norm": 0.7474587559700012, "learning_rate": 0.0002, "epoch": 3.135428793170353, "step": 4040}, {"loss": 1.4616, "grad_norm": 0.7292373776435852, "learning_rate": 0.0002, "epoch": 3.1431897555296855, "step": 4050}, {"loss": 1.3908, "grad_norm": 0.7432886958122253, "learning_rate": 0.0002, "epoch": 3.1509507178890184, "step": 4060}, {"loss": 1.4214, "grad_norm": 0.6366098523139954, "learning_rate": 0.0002, "epoch": 3.158711680248351, "step": 4070}, {"loss": 1.5044, "grad_norm": 0.6837611794471741, "learning_rate": 0.0002, "epoch": 3.1664726426076832, "step": 4080}, {"loss": 1.4332, "grad_norm": 0.7194393277168274, "learning_rate": 0.0002, "epoch": 3.174233604967016, "step": 4090}, {"loss": 1.3628, "grad_norm": 0.6963607668876648, "learning_rate": 0.0002, "epoch": 3.1819945673263486, "step": 4100}, {"loss": 1.4127, "grad_norm": 0.6404902935028076, "learning_rate": 0.0002, "epoch": 3.189755529685681, "step": 4110}, {"loss": 1.4394, "grad_norm": 0.7172070741653442, "learning_rate": 0.0002, "epoch": 3.1975164920450134, "step": 4120}, {"loss": 1.4658, "grad_norm": 0.6577759385108948, "learning_rate": 0.0002, "epoch": 3.2052774544043463, "step": 4130}, {"loss": 1.4019, "grad_norm": 0.6658480167388916, "learning_rate": 0.0002, "epoch": 3.2130384167636787, "step": 4140}, {"loss": 1.4348, "grad_norm": 0.6771699786186218, "learning_rate": 0.0002, "epoch": 3.220799379123011, "step": 4150}, {"loss": 1.4736, "grad_norm": 0.699035108089447, "learning_rate": 0.0002, "epoch": 3.2285603414823436, "step": 4160}, {"loss": 1.4096, "grad_norm": 0.7218514680862427, "learning_rate": 0.0002, "epoch": 3.2363213038416765, "step": 4170}, {"loss": 1.3637, "grad_norm": 0.6270631551742554, "learning_rate": 0.0002, "epoch": 3.244082266201009, "step": 4180}, {"loss": 1.4076, "grad_norm": 0.6828921437263489, "learning_rate": 0.0002, "epoch": 3.2518432285603414, "step": 4190}, {"loss": 1.4663, "grad_norm": 0.6005498170852661, "learning_rate": 0.0002, "epoch": 3.2596041909196742, "step": 4200}, {"loss": 1.4798, "grad_norm": 0.6974790692329407, "learning_rate": 0.0002, "epoch": 3.2673651532790067, "step": 4210}, {"loss": 1.5012, "grad_norm": 0.7269543409347534, "learning_rate": 0.0002, "epoch": 3.275126115638339, "step": 4220}, {"loss": 1.3848, "grad_norm": 0.6728787422180176, "learning_rate": 0.0002, "epoch": 3.2828870779976715, "step": 4230}, {"loss": 1.4112, "grad_norm": 0.676972508430481, "learning_rate": 0.0002, "epoch": 3.2906480403570044, "step": 4240}, {"loss": 1.4206, "grad_norm": 0.748309314250946, "learning_rate": 0.0002, "epoch": 3.298409002716337, "step": 4250}, {"loss": 1.4973, "grad_norm": 0.6976589560508728, "learning_rate": 0.0002, "epoch": 3.3061699650756693, "step": 4260}, {"loss": 1.3967, "grad_norm": 0.649780809879303, "learning_rate": 0.0002, "epoch": 3.3139309274350017, "step": 4270}, {"loss": 1.327, "grad_norm": 0.6529902815818787, "learning_rate": 0.0002, "epoch": 3.3216918897943346, "step": 4280}, {"loss": 1.4888, "grad_norm": 0.9273163676261902, "learning_rate": 0.0002, "epoch": 3.329452852153667, "step": 4290}, {"loss": 1.4859, "grad_norm": 0.717024028301239, "learning_rate": 0.0002, "epoch": 3.3372138145129995, "step": 4300}, {"loss": 1.4441, "grad_norm": 0.7914950251579285, "learning_rate": 0.0002, "epoch": 3.3449747768723324, "step": 4310}, {"loss": 1.432, "grad_norm": 0.7133203148841858, "learning_rate": 0.0002, "epoch": 3.352735739231665, "step": 4320}, {"loss": 1.4662, "grad_norm": 0.7409568428993225, "learning_rate": 0.0002, "epoch": 3.3604967015909972, "step": 4330}, {"loss": 1.3992, "grad_norm": 0.6993981003761292, "learning_rate": 0.0002, "epoch": 3.3682576639503297, "step": 4340}, {"loss": 1.4261, "grad_norm": 0.7114535570144653, "learning_rate": 0.0002, "epoch": 3.3760186263096625, "step": 4350}, {"loss": 1.4227, "grad_norm": 0.6790860295295715, "learning_rate": 0.0002, "epoch": 3.383779588668995, "step": 4360}, {"loss": 1.4128, "grad_norm": 0.6507849097251892, "learning_rate": 0.0002, "epoch": 3.3915405510283274, "step": 4370}, {"loss": 1.4559, "grad_norm": 0.5967804193496704, "learning_rate": 0.0002, "epoch": 3.39930151338766, "step": 4380}, {"loss": 1.3687, "grad_norm": 0.6625847816467285, "learning_rate": 0.0002, "epoch": 3.4070624757469927, "step": 4390}, {"loss": 1.4193, "grad_norm": 0.6736508011817932, "learning_rate": 0.0002, "epoch": 3.414823438106325, "step": 4400}, {"loss": 1.4363, "grad_norm": 0.7870860695838928, "learning_rate": 0.0002, "epoch": 3.4225844004656576, "step": 4410}, {"loss": 1.4114, "grad_norm": 0.7205295562744141, "learning_rate": 0.0002, "epoch": 3.4303453628249905, "step": 4420}, {"loss": 1.4131, "grad_norm": 0.6634634137153625, "learning_rate": 0.0002, "epoch": 3.438106325184323, "step": 4430}, {"loss": 1.4683, "grad_norm": 0.7562733292579651, "learning_rate": 0.0002, "epoch": 3.4458672875436553, "step": 4440}, {"loss": 1.3486, "grad_norm": 0.6585879921913147, "learning_rate": 0.0002, "epoch": 3.453628249902988, "step": 4450}, {"loss": 1.4283, "grad_norm": 0.6896792054176331, "learning_rate": 0.0002, "epoch": 3.4613892122623207, "step": 4460}, {"loss": 1.4208, "grad_norm": 0.6520342230796814, "learning_rate": 0.0002, "epoch": 3.469150174621653, "step": 4470}, {"loss": 1.3423, "grad_norm": 0.6760806441307068, "learning_rate": 0.0002, "epoch": 3.4769111369809855, "step": 4480}, {"loss": 1.4398, "grad_norm": 0.7539774179458618, "learning_rate": 0.0002, "epoch": 3.484672099340318, "step": 4490}, {"loss": 1.4534, "grad_norm": 0.7409411668777466, "learning_rate": 0.0002, "epoch": 3.492433061699651, "step": 4500}, {"loss": 1.4069, "grad_norm": 0.6876253485679626, "learning_rate": 0.0002, "epoch": 3.5001940240589833, "step": 4510}, {"loss": 1.4228, "grad_norm": 0.7028461694717407, "learning_rate": 0.0002, "epoch": 3.5079549864183157, "step": 4520}, {"loss": 1.4723, "grad_norm": 0.8056529760360718, "learning_rate": 0.0002, "epoch": 3.5157159487776486, "step": 4530}, {"loss": 1.4148, "grad_norm": 0.711338996887207, "learning_rate": 0.0002, "epoch": 3.523476911136981, "step": 4540}, {"loss": 1.5247, "grad_norm": 0.7343552708625793, "learning_rate": 0.0002, "epoch": 3.5312378734963135, "step": 4550}, {"loss": 1.4308, "grad_norm": 0.745479941368103, "learning_rate": 0.0002, "epoch": 3.5389988358556463, "step": 4560}, {"loss": 1.4229, "grad_norm": 0.7582294940948486, "learning_rate": 0.0002, "epoch": 3.5467597982149788, "step": 4570}, {"loss": 1.4127, "grad_norm": 0.6717444658279419, "learning_rate": 0.0002, "epoch": 3.554520760574311, "step": 4580}, {"loss": 1.4368, "grad_norm": 0.7417883276939392, "learning_rate": 0.0002, "epoch": 3.5622817229336436, "step": 4590}, {"loss": 1.4176, "grad_norm": 0.6385737061500549, "learning_rate": 0.0002, "epoch": 3.570042685292976, "step": 4600}, {"loss": 1.3981, "grad_norm": 0.716704249382019, "learning_rate": 0.0002, "epoch": 3.577803647652309, "step": 4610}, {"loss": 1.3889, "grad_norm": 0.6948980093002319, "learning_rate": 0.0002, "epoch": 3.5855646100116414, "step": 4620}, {"loss": 1.5177, "grad_norm": 0.6961140036582947, "learning_rate": 0.0002, "epoch": 3.593325572370974, "step": 4630}, {"loss": 1.4508, "grad_norm": 0.7493122220039368, "learning_rate": 0.0002, "epoch": 3.6010865347303067, "step": 4640}, {"loss": 1.3987, "grad_norm": 0.7431658506393433, "learning_rate": 0.0002, "epoch": 3.608847497089639, "step": 4650}, {"loss": 1.4551, "grad_norm": 0.8353387713432312, "learning_rate": 0.0002, "epoch": 3.6166084594489716, "step": 4660}, {"loss": 1.4533, "grad_norm": 0.7095612287521362, "learning_rate": 0.0002, "epoch": 3.6243694218083045, "step": 4670}, {"loss": 1.4003, "grad_norm": 0.776620090007782, "learning_rate": 0.0002, "epoch": 3.632130384167637, "step": 4680}, {"loss": 1.4361, "grad_norm": 0.7198925018310547, "learning_rate": 0.0002, "epoch": 3.6398913465269693, "step": 4690}, {"loss": 1.4543, "grad_norm": 0.8238834738731384, "learning_rate": 0.0002, "epoch": 3.6476523088863018, "step": 4700}, {"loss": 1.3958, "grad_norm": 0.6804245710372925, "learning_rate": 0.0002, "epoch": 3.655413271245634, "step": 4710}, {"loss": 1.4158, "grad_norm": 0.8444845676422119, "learning_rate": 0.0002, "epoch": 3.663174233604967, "step": 4720}, {"loss": 1.3825, "grad_norm": 0.743797779083252, "learning_rate": 0.0002, "epoch": 3.6709351959642995, "step": 4730}, {"loss": 1.4213, "grad_norm": 0.8994188904762268, "learning_rate": 0.0002, "epoch": 3.678696158323632, "step": 4740}, {"loss": 1.4281, "grad_norm": 0.75416100025177, "learning_rate": 0.0002, "epoch": 3.686457120682965, "step": 4750}, {"loss": 1.4154, "grad_norm": 0.6499266028404236, "learning_rate": 0.0002, "epoch": 3.6942180830422973, "step": 4760}, {"loss": 1.4005, "grad_norm": 0.7246791124343872, "learning_rate": 0.0002, "epoch": 3.7019790454016297, "step": 4770}, {"loss": 1.426, "grad_norm": 0.7831124067306519, "learning_rate": 0.0002, "epoch": 3.7097400077609626, "step": 4780}, {"loss": 1.3933, "grad_norm": 0.7130028009414673, "learning_rate": 0.0002, "epoch": 3.717500970120295, "step": 4790}, {"loss": 1.4632, "grad_norm": 0.7501602172851562, "learning_rate": 0.0002, "epoch": 3.7252619324796274, "step": 4800}, {"loss": 1.4985, "grad_norm": 0.6980932950973511, "learning_rate": 0.0002, "epoch": 3.73302289483896, "step": 4810}, {"loss": 1.4517, "grad_norm": 0.8050530552864075, "learning_rate": 0.0002, "epoch": 3.7407838571982923, "step": 4820}, {"loss": 1.4703, "grad_norm": 0.6385579705238342, "learning_rate": 0.0002, "epoch": 3.748544819557625, "step": 4830}, {"loss": 1.5281, "grad_norm": 0.6664714813232422, "learning_rate": 0.0002, "epoch": 3.7563057819169576, "step": 4840}, {"loss": 1.4443, "grad_norm": 0.7125676274299622, "learning_rate": 0.0002, "epoch": 3.76406674427629, "step": 4850}, {"loss": 1.3958, "grad_norm": 0.7231866717338562, "learning_rate": 0.0002, "epoch": 3.771827706635623, "step": 4860}, {"loss": 1.4446, "grad_norm": 0.6917183995246887, "learning_rate": 0.0002, "epoch": 3.7795886689949554, "step": 4870}, {"loss": 1.4369, "grad_norm": 0.665037989616394, "learning_rate": 0.0002, "epoch": 3.787349631354288, "step": 4880}, {"loss": 1.4193, "grad_norm": 0.5837726593017578, "learning_rate": 0.0002, "epoch": 3.7951105937136207, "step": 4890}, {"loss": 1.4176, "grad_norm": 0.6366701722145081, "learning_rate": 0.0002, "epoch": 3.802871556072953, "step": 4900}, {"loss": 1.46, "grad_norm": 0.7082223892211914, "learning_rate": 0.0002, "epoch": 3.8106325184322856, "step": 4910}, {"loss": 1.5139, "grad_norm": 0.8101672530174255, "learning_rate": 0.0002, "epoch": 3.818393480791618, "step": 4920}, {"loss": 1.3659, "grad_norm": 0.7516148090362549, "learning_rate": 0.0002, "epoch": 3.826154443150951, "step": 4930}, {"loss": 1.3909, "grad_norm": 0.7928489446640015, "learning_rate": 0.0002, "epoch": 3.8339154055102833, "step": 4940}, {"loss": 1.4255, "grad_norm": 0.6892234683036804, "learning_rate": 0.0002, "epoch": 3.8416763678696157, "step": 4950}, {"loss": 1.5024, "grad_norm": 0.6381304264068604, "learning_rate": 0.0002, "epoch": 3.849437330228948, "step": 4960}, {"loss": 1.4873, "grad_norm": 0.8068831562995911, "learning_rate": 0.0002, "epoch": 3.857198292588281, "step": 4970}, {"loss": 1.45, "grad_norm": 0.7289869785308838, "learning_rate": 0.0002, "epoch": 3.8649592549476135, "step": 4980}, {"loss": 1.398, "grad_norm": 0.7278549075126648, "learning_rate": 0.0002, "epoch": 3.872720217306946, "step": 4990}, {"loss": 1.4442, "grad_norm": 0.7324236631393433, "learning_rate": 0.0002, "epoch": 3.880481179666279, "step": 5000}, {"loss": 1.4511, "grad_norm": 0.6759871244430542, "learning_rate": 0.0002, "epoch": 3.8882421420256112, "step": 5010}, {"loss": 1.4705, "grad_norm": 0.8159207701683044, "learning_rate": 0.0002, "epoch": 3.8960031043849437, "step": 5020}, {"loss": 1.4685, "grad_norm": 0.6536211967468262, "learning_rate": 0.0002, "epoch": 3.9037640667442766, "step": 5030}, {"loss": 1.4335, "grad_norm": 0.6827932000160217, "learning_rate": 0.0002, "epoch": 3.911525029103609, "step": 5040}, {"loss": 1.433, "grad_norm": 0.6688340306282043, "learning_rate": 0.0002, "epoch": 3.9192859914629414, "step": 5050}, {"loss": 1.4099, "grad_norm": 0.6385695934295654, "learning_rate": 0.0002, "epoch": 3.927046953822274, "step": 5060}, {"loss": 1.4767, "grad_norm": 0.6975107192993164, "learning_rate": 0.0002, "epoch": 3.9348079161816063, "step": 5070}, {"loss": 1.4893, "grad_norm": 0.6684112548828125, "learning_rate": 0.0002, "epoch": 3.942568878540939, "step": 5080}, {"loss": 1.4732, "grad_norm": 0.8349628448486328, "learning_rate": 0.0002, "epoch": 3.9503298409002716, "step": 5090}, {"loss": 1.5131, "grad_norm": 0.7146425843238831, "learning_rate": 0.0002, "epoch": 3.958090803259604, "step": 5100}, {"loss": 1.4149, "grad_norm": 0.6555036902427673, "learning_rate": 0.0002, "epoch": 3.965851765618937, "step": 5110}, {"loss": 1.4274, "grad_norm": 0.7037415504455566, "learning_rate": 0.0002, "epoch": 3.9736127279782694, "step": 5120}, {"loss": 1.4292, "grad_norm": 0.7235575914382935, "learning_rate": 0.0002, "epoch": 3.981373690337602, "step": 5130}, {"loss": 1.4455, "grad_norm": 0.7092325687408447, "learning_rate": 0.0002, "epoch": 3.9891346526969347, "step": 5140}, {"loss": 1.4512, "grad_norm": 0.7490319609642029, "learning_rate": 0.0002, "epoch": 3.996895615056267, "step": 5150}, {"eval_loss": 1.9131355285644531, "eval_runtime": 105.5778, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 4.0, "step": 5154}, {"loss": 1.2643, "grad_norm": 0.7075854539871216, "learning_rate": 0.0002, "epoch": 4.0046565774155995, "step": 5160}, {"loss": 1.209, "grad_norm": 0.9466007351875305, "learning_rate": 0.0002, "epoch": 4.012417539774932, "step": 5170}, {"loss": 1.2567, "grad_norm": 1.0297044515609741, "learning_rate": 0.0002, "epoch": 4.020178502134264, "step": 5180}, {"loss": 1.1796, "grad_norm": 0.7765059471130371, "learning_rate": 0.0002, "epoch": 4.027939464493597, "step": 5190}, {"loss": 1.2356, "grad_norm": 0.995760977268219, "learning_rate": 0.0002, "epoch": 4.03570042685293, "step": 5200}, {"loss": 1.1792, "grad_norm": 0.8663829565048218, "learning_rate": 0.0002, "epoch": 4.043461389212262, "step": 5210}, {"loss": 1.2471, "grad_norm": 1.0660825967788696, "learning_rate": 0.0002, "epoch": 4.051222351571595, "step": 5220}, {"loss": 1.1676, "grad_norm": 0.9858174920082092, "learning_rate": 0.0002, "epoch": 4.058983313930927, "step": 5230}, {"loss": 1.2448, "grad_norm": 0.8911338448524475, "learning_rate": 0.0002, "epoch": 4.06674427629026, "step": 5240}, {"loss": 1.1858, "grad_norm": 1.0848394632339478, "learning_rate": 0.0002, "epoch": 4.074505238649593, "step": 5250}, {"loss": 1.1684, "grad_norm": 1.0849905014038086, "learning_rate": 0.0002, "epoch": 4.082266201008925, "step": 5260}, {"loss": 1.2007, "grad_norm": 1.0497841835021973, "learning_rate": 0.0002, "epoch": 4.090027163368258, "step": 5270}, {"loss": 1.2552, "grad_norm": 0.8943053483963013, "learning_rate": 0.0002, "epoch": 4.0977881257275905, "step": 5280}, {"loss": 1.1923, "grad_norm": 0.8432527184486389, "learning_rate": 0.0002, "epoch": 4.1055490880869225, "step": 5290}, {"loss": 1.1634, "grad_norm": 0.9690414667129517, "learning_rate": 0.0002, "epoch": 4.113310050446255, "step": 5300}, {"loss": 1.3019, "grad_norm": 0.7790773510932922, "learning_rate": 0.0002, "epoch": 4.121071012805588, "step": 5310}, {"loss": 1.1806, "grad_norm": 0.9289211630821228, "learning_rate": 0.0002, "epoch": 4.12883197516492, "step": 5320}, {"loss": 1.1458, "grad_norm": 1.0785125494003296, "learning_rate": 0.0002, "epoch": 4.136592937524253, "step": 5330}, {"loss": 1.2086, "grad_norm": 0.8559591770172119, "learning_rate": 0.0002, "epoch": 4.144353899883585, "step": 5340}, {"loss": 1.1974, "grad_norm": 0.9405956268310547, "learning_rate": 0.0002, "epoch": 4.152114862242918, "step": 5350}, {"loss": 1.1793, "grad_norm": 0.9942827820777893, "learning_rate": 0.0002, "epoch": 4.159875824602251, "step": 5360}, {"loss": 1.1659, "grad_norm": 0.9141933917999268, "learning_rate": 0.0002, "epoch": 4.167636786961583, "step": 5370}, {"loss": 1.1647, "grad_norm": 0.8206015229225159, "learning_rate": 0.0002, "epoch": 4.175397749320916, "step": 5380}, {"loss": 1.2778, "grad_norm": 0.9340888857841492, "learning_rate": 0.0002, "epoch": 4.183158711680249, "step": 5390}, {"loss": 1.2459, "grad_norm": 1.2122114896774292, "learning_rate": 0.0002, "epoch": 4.190919674039581, "step": 5400}, {"loss": 1.2371, "grad_norm": 1.0661298036575317, "learning_rate": 0.0002, "epoch": 4.1986806363989135, "step": 5410}, {"loss": 1.1978, "grad_norm": 0.9372861385345459, "learning_rate": 0.0002, "epoch": 4.206441598758246, "step": 5420}, {"loss": 1.2653, "grad_norm": 0.894012987613678, "learning_rate": 0.0002, "epoch": 4.214202561117578, "step": 5430}, {"loss": 1.387, "grad_norm": 1.0647753477096558, "learning_rate": 0.0002, "epoch": 4.221963523476911, "step": 5440}, {"loss": 1.2231, "grad_norm": 0.989179790019989, "learning_rate": 0.0002, "epoch": 4.229724485836243, "step": 5450}, {"loss": 1.2715, "grad_norm": 1.1601181030273438, "learning_rate": 0.0002, "epoch": 4.237485448195576, "step": 5460}, {"loss": 1.2406, "grad_norm": 0.9395585656166077, "learning_rate": 0.0002, "epoch": 4.245246410554909, "step": 5470}, {"loss": 1.2779, "grad_norm": 0.9527766108512878, "learning_rate": 0.0002, "epoch": 4.253007372914241, "step": 5480}, {"loss": 1.267, "grad_norm": 1.0319520235061646, "learning_rate": 0.0002, "epoch": 4.260768335273574, "step": 5490}, {"loss": 1.2633, "grad_norm": 0.8659824728965759, "learning_rate": 0.0002, "epoch": 4.268529297632907, "step": 5500}, {"loss": 1.1475, "grad_norm": 1.099211573600769, "learning_rate": 0.0002, "epoch": 4.276290259992239, "step": 5510}, {"loss": 1.2508, "grad_norm": 0.9363361597061157, "learning_rate": 0.0002, "epoch": 4.284051222351572, "step": 5520}, {"loss": 1.189, "grad_norm": 0.8437647223472595, "learning_rate": 0.0002, "epoch": 4.2918121847109045, "step": 5530}, {"loss": 1.2212, "grad_norm": 0.9181258678436279, "learning_rate": 0.0002, "epoch": 4.2995731470702365, "step": 5540}, {"loss": 1.2092, "grad_norm": 0.9059357643127441, "learning_rate": 0.0002, "epoch": 4.307334109429569, "step": 5550}, {"loss": 1.2189, "grad_norm": 0.9337241649627686, "learning_rate": 0.0002, "epoch": 4.315095071788901, "step": 5560}, {"loss": 1.2462, "grad_norm": 0.9428889155387878, "learning_rate": 0.0002, "epoch": 4.322856034148234, "step": 5570}, {"loss": 1.2675, "grad_norm": 1.003589153289795, "learning_rate": 0.0002, "epoch": 4.330616996507567, "step": 5580}, {"loss": 1.2703, "grad_norm": 1.1249268054962158, "learning_rate": 0.0002, "epoch": 4.338377958866899, "step": 5590}, {"loss": 1.2501, "grad_norm": 0.8623469471931458, "learning_rate": 0.0002, "epoch": 4.346138921226232, "step": 5600}, {"loss": 1.2404, "grad_norm": 1.1389174461364746, "learning_rate": 0.0002, "epoch": 4.353899883585565, "step": 5610}, {"loss": 1.2245, "grad_norm": 1.0136264562606812, "learning_rate": 0.0002, "epoch": 4.361660845944897, "step": 5620}, {"loss": 1.3473, "grad_norm": 0.9567070603370667, "learning_rate": 0.0002, "epoch": 4.36942180830423, "step": 5630}, {"loss": 1.2988, "grad_norm": 1.0592148303985596, "learning_rate": 0.0002, "epoch": 4.377182770663563, "step": 5640}, {"loss": 1.212, "grad_norm": 1.0110485553741455, "learning_rate": 0.0002, "epoch": 4.384943733022895, "step": 5650}, {"loss": 1.2086, "grad_norm": 0.9914907217025757, "learning_rate": 0.0002, "epoch": 4.3927046953822275, "step": 5660}, {"loss": 1.2363, "grad_norm": 0.9447247982025146, "learning_rate": 0.0002, "epoch": 4.4004656577415595, "step": 5670}, {"loss": 1.2617, "grad_norm": 0.9644378423690796, "learning_rate": 0.0002, "epoch": 4.408226620100892, "step": 5680}, {"loss": 1.2773, "grad_norm": 0.920676589012146, "learning_rate": 0.0002, "epoch": 4.415987582460225, "step": 5690}, {"loss": 1.2792, "grad_norm": 1.060570478439331, "learning_rate": 0.0002, "epoch": 4.423748544819557, "step": 5700}, {"loss": 1.2374, "grad_norm": 0.8857738971710205, "learning_rate": 0.0002, "epoch": 4.43150950717889, "step": 5710}, {"loss": 1.2588, "grad_norm": 1.0536398887634277, "learning_rate": 0.0002, "epoch": 4.439270469538223, "step": 5720}, {"loss": 1.2051, "grad_norm": 0.990847110748291, "learning_rate": 0.0002, "epoch": 4.447031431897555, "step": 5730}, {"loss": 1.2469, "grad_norm": 0.9692499041557312, "learning_rate": 0.0002, "epoch": 4.454792394256888, "step": 5740}, {"loss": 1.2269, "grad_norm": 1.0376402139663696, "learning_rate": 0.0002, "epoch": 4.462553356616221, "step": 5750}, {"loss": 1.1701, "grad_norm": 1.3863259553909302, "learning_rate": 0.0002, "epoch": 4.470314318975553, "step": 5760}, {"loss": 1.2591, "grad_norm": 0.978379487991333, "learning_rate": 0.0002, "epoch": 4.478075281334886, "step": 5770}, {"loss": 1.2729, "grad_norm": 1.0973085165023804, "learning_rate": 0.0002, "epoch": 4.485836243694218, "step": 5780}, {"loss": 1.2404, "grad_norm": 1.057006597518921, "learning_rate": 0.0002, "epoch": 4.4935972060535505, "step": 5790}, {"loss": 1.2476, "grad_norm": 0.9247729182243347, "learning_rate": 0.0002, "epoch": 4.501358168412883, "step": 5800}, {"loss": 1.2369, "grad_norm": 1.0447787046432495, "learning_rate": 0.0002, "epoch": 4.509119130772215, "step": 5810}, {"loss": 1.211, "grad_norm": 1.1930429935455322, "learning_rate": 0.0002, "epoch": 4.516880093131548, "step": 5820}, {"loss": 1.2596, "grad_norm": 0.9867590069770813, "learning_rate": 0.0002, "epoch": 4.524641055490881, "step": 5830}, {"loss": 1.2766, "grad_norm": 0.9591100215911865, "learning_rate": 0.0002, "epoch": 4.532402017850213, "step": 5840}, {"loss": 1.2154, "grad_norm": 0.9950753450393677, "learning_rate": 0.0002, "epoch": 4.540162980209546, "step": 5850}, {"loss": 1.2149, "grad_norm": 1.0087506771087646, "learning_rate": 0.0002, "epoch": 4.547923942568879, "step": 5860}, {"loss": 1.3165, "grad_norm": 1.0934417247772217, "learning_rate": 0.0002, "epoch": 4.555684904928211, "step": 5870}, {"loss": 1.3059, "grad_norm": 1.107987403869629, "learning_rate": 0.0002, "epoch": 4.563445867287544, "step": 5880}, {"loss": 1.2184, "grad_norm": 0.9147276878356934, "learning_rate": 0.0002, "epoch": 4.571206829646876, "step": 5890}, {"loss": 1.24, "grad_norm": 1.036780595779419, "learning_rate": 0.0002, "epoch": 4.578967792006209, "step": 5900}, {"loss": 1.2209, "grad_norm": 0.9284719824790955, "learning_rate": 0.0002, "epoch": 4.5867287543655415, "step": 5910}, {"loss": 1.3693, "grad_norm": 0.9141898155212402, "learning_rate": 0.0002, "epoch": 4.5944897167248735, "step": 5920}, {"loss": 1.2319, "grad_norm": 1.0447357892990112, "learning_rate": 0.0002, "epoch": 4.602250679084206, "step": 5930}, {"loss": 1.2667, "grad_norm": 0.9309114217758179, "learning_rate": 0.0002, "epoch": 4.610011641443539, "step": 5940}, {"loss": 1.2827, "grad_norm": 1.2986129522323608, "learning_rate": 0.0002, "epoch": 4.617772603802871, "step": 5950}, {"loss": 1.312, "grad_norm": 0.9221704602241516, "learning_rate": 0.0002, "epoch": 4.625533566162204, "step": 5960}, {"loss": 1.2769, "grad_norm": 0.9228187799453735, "learning_rate": 0.0002, "epoch": 4.633294528521537, "step": 5970}, {"loss": 1.2953, "grad_norm": 0.9483116269111633, "learning_rate": 0.0002, "epoch": 4.641055490880869, "step": 5980}, {"loss": 1.3437, "grad_norm": 1.0218974351882935, "learning_rate": 0.0002, "epoch": 4.648816453240202, "step": 5990}, {"loss": 1.3085, "grad_norm": 0.9764600396156311, "learning_rate": 0.0002, "epoch": 4.656577415599534, "step": 6000}, {"loss": 1.197, "grad_norm": 0.9115710258483887, "learning_rate": 0.0002, "epoch": 4.664338377958867, "step": 6010}, {"loss": 1.1917, "grad_norm": 0.9245651364326477, "learning_rate": 0.0002, "epoch": 4.6720993403182, "step": 6020}, {"loss": 1.2969, "grad_norm": 0.9686311483383179, "learning_rate": 0.0002, "epoch": 4.6798603026775325, "step": 6030}, {"loss": 1.2702, "grad_norm": 1.1807392835617065, "learning_rate": 0.0002, "epoch": 4.6876212650368645, "step": 6040}, {"loss": 1.328, "grad_norm": 1.0358641147613525, "learning_rate": 0.0002, "epoch": 4.695382227396197, "step": 6050}, {"loss": 1.3281, "grad_norm": 0.987332284450531, "learning_rate": 0.0002, "epoch": 4.703143189755529, "step": 6060}, {"loss": 1.2514, "grad_norm": 1.0526494979858398, "learning_rate": 0.0002, "epoch": 4.710904152114862, "step": 6070}, {"loss": 1.2246, "grad_norm": 1.0276758670806885, "learning_rate": 0.0002, "epoch": 4.718665114474195, "step": 6080}, {"loss": 1.3367, "grad_norm": 0.9904406666755676, "learning_rate": 0.0002, "epoch": 4.726426076833527, "step": 6090}, {"loss": 1.2797, "grad_norm": 1.0084882974624634, "learning_rate": 0.0002, "epoch": 4.73418703919286, "step": 6100}, {"loss": 1.2656, "grad_norm": 0.8646450638771057, "learning_rate": 0.0002, "epoch": 4.741948001552192, "step": 6110}, {"loss": 1.3063, "grad_norm": 0.9233377575874329, "learning_rate": 0.0002, "epoch": 4.749708963911525, "step": 6120}, {"loss": 1.2642, "grad_norm": 0.9675140976905823, "learning_rate": 0.0002, "epoch": 4.757469926270858, "step": 6130}, {"loss": 1.3367, "grad_norm": 0.9639796018600464, "learning_rate": 0.0002, "epoch": 4.765230888630191, "step": 6140}, {"loss": 1.276, "grad_norm": 0.925199568271637, "learning_rate": 0.0002, "epoch": 4.772991850989523, "step": 6150}, {"loss": 1.2441, "grad_norm": 1.050901174545288, "learning_rate": 0.0002, "epoch": 4.7807528133488555, "step": 6160}, {"loss": 1.301, "grad_norm": 0.8920623660087585, "learning_rate": 0.0002, "epoch": 4.7885137757081875, "step": 6170}, {"loss": 1.263, "grad_norm": 0.8964757919311523, "learning_rate": 0.0002, "epoch": 4.79627473806752, "step": 6180}, {"loss": 1.2787, "grad_norm": 1.0839070081710815, "learning_rate": 0.0002, "epoch": 4.804035700426853, "step": 6190}, {"loss": 1.2664, "grad_norm": 0.8809942007064819, "learning_rate": 0.0002, "epoch": 4.811796662786185, "step": 6200}, {"loss": 1.321, "grad_norm": 1.0216195583343506, "learning_rate": 0.0002, "epoch": 4.819557625145518, "step": 6210}, {"loss": 1.3033, "grad_norm": 0.892005980014801, "learning_rate": 0.0002, "epoch": 4.827318587504851, "step": 6220}, {"loss": 1.2602, "grad_norm": 0.9957166910171509, "learning_rate": 0.0002, "epoch": 4.835079549864183, "step": 6230}, {"loss": 1.3562, "grad_norm": 0.9720533490180969, "learning_rate": 0.0002, "epoch": 4.842840512223516, "step": 6240}, {"loss": 1.2651, "grad_norm": 0.9336182475090027, "learning_rate": 0.0002, "epoch": 4.850601474582849, "step": 6250}, {"loss": 1.3136, "grad_norm": 1.2611457109451294, "learning_rate": 0.0002, "epoch": 4.858362436942181, "step": 6260}, {"loss": 1.2234, "grad_norm": 0.8927203416824341, "learning_rate": 0.0002, "epoch": 4.866123399301514, "step": 6270}, {"loss": 1.3463, "grad_norm": 0.9706710577011108, "learning_rate": 0.0002, "epoch": 4.873884361660846, "step": 6280}, {"loss": 1.3209, "grad_norm": 1.1461690664291382, "learning_rate": 0.0002, "epoch": 4.8816453240201785, "step": 6290}, {"loss": 1.2566, "grad_norm": 0.9930381178855896, "learning_rate": 0.0002, "epoch": 4.889406286379511, "step": 6300}, {"loss": 1.2568, "grad_norm": 0.91451096534729, "learning_rate": 0.0002, "epoch": 4.897167248738843, "step": 6310}, {"loss": 1.2836, "grad_norm": 1.0319571495056152, "learning_rate": 0.0002, "epoch": 4.904928211098176, "step": 6320}, {"loss": 1.2908, "grad_norm": 0.990140438079834, "learning_rate": 0.0002, "epoch": 4.912689173457509, "step": 6330}, {"loss": 1.3299, "grad_norm": 1.2466117143630981, "learning_rate": 0.0002, "epoch": 4.920450135816841, "step": 6340}, {"loss": 1.2659, "grad_norm": 1.0316979885101318, "learning_rate": 0.0002, "epoch": 4.928211098176174, "step": 6350}, {"loss": 1.3292, "grad_norm": 1.0643759965896606, "learning_rate": 0.0002, "epoch": 4.935972060535507, "step": 6360}, {"loss": 1.2559, "grad_norm": 0.9703279733657837, "learning_rate": 0.0002, "epoch": 4.943733022894839, "step": 6370}, {"loss": 1.2155, "grad_norm": 0.9767927527427673, "learning_rate": 0.0002, "epoch": 4.951493985254172, "step": 6380}, {"loss": 1.2437, "grad_norm": 0.960854172706604, "learning_rate": 0.0002, "epoch": 4.959254947613504, "step": 6390}, {"loss": 1.3314, "grad_norm": 0.9922910332679749, "learning_rate": 0.0002, "epoch": 4.967015909972837, "step": 6400}, {"loss": 1.3018, "grad_norm": 0.956470787525177, "learning_rate": 0.0002, "epoch": 4.9747768723321695, "step": 6410}, {"loss": 1.2794, "grad_norm": 0.9637242555618286, "learning_rate": 0.0002, "epoch": 4.9825378346915015, "step": 6420}, {"loss": 1.3236, "grad_norm": 1.0855202674865723, "learning_rate": 0.0002, "epoch": 4.990298797050834, "step": 6430}, {"loss": 1.3015, "grad_norm": 0.9655316472053528, "learning_rate": 0.0002, "epoch": 4.998059759410167, "step": 6440}, {"eval_loss": 2.0410802364349365, "eval_runtime": 113.04, "eval_samples_per_second": 4.485, "eval_steps_per_second": 0.566, "epoch": 4.9996119518820334, "step": 6442}, {"loss": 1.0846, "grad_norm": 1.1676199436187744, "learning_rate": 0.0002, "epoch": 5.005820721769499, "step": 6450}, {"loss": 1.041, "grad_norm": 1.4317965507507324, "learning_rate": 0.0002, "epoch": 5.013581684128832, "step": 6460}, {"loss": 0.9546, "grad_norm": 1.460443377494812, "learning_rate": 0.0002, "epoch": 5.021342646488165, "step": 6470}, {"loss": 1.0014, "grad_norm": 1.2299214601516724, "learning_rate": 0.0002, "epoch": 5.029103608847497, "step": 6480}, {"loss": 1.0397, "grad_norm": 1.3125724792480469, "learning_rate": 0.0002, "epoch": 5.03686457120683, "step": 6490}, {"loss": 1.0134, "grad_norm": 1.1252319812774658, "learning_rate": 0.0002, "epoch": 5.044625533566162, "step": 6500}, {"loss": 0.976, "grad_norm": 0.9970866441726685, "learning_rate": 0.0002, "epoch": 5.052386495925495, "step": 6510}, {"loss": 0.9731, "grad_norm": 1.229069709777832, "learning_rate": 0.0002, "epoch": 5.060147458284828, "step": 6520}, {"loss": 1.0498, "grad_norm": 1.2430938482284546, "learning_rate": 0.0002, "epoch": 5.06790842064416, "step": 6530}, {"loss": 1.0236, "grad_norm": 1.0522737503051758, "learning_rate": 0.0002, "epoch": 5.0756693830034925, "step": 6540}, {"loss": 1.0221, "grad_norm": 1.108890175819397, "learning_rate": 0.0002, "epoch": 5.083430345362825, "step": 6550}, {"loss": 1.0177, "grad_norm": 1.156912922859192, "learning_rate": 0.0002, "epoch": 5.091191307722157, "step": 6560}, {"loss": 1.0415, "grad_norm": 1.405895709991455, "learning_rate": 0.0002, "epoch": 5.09895227008149, "step": 6570}, {"loss": 0.9811, "grad_norm": 1.2005155086517334, "learning_rate": 0.0002, "epoch": 5.106713232440823, "step": 6580}, {"loss": 0.9862, "grad_norm": 1.181443452835083, "learning_rate": 0.0002, "epoch": 5.114474194800155, "step": 6590}, {"loss": 1.0291, "grad_norm": 2.3444771766662598, "learning_rate": 0.0002, "epoch": 5.122235157159488, "step": 6600}, {"loss": 1.0455, "grad_norm": 1.216988444328308, "learning_rate": 0.0002, "epoch": 5.12999611951882, "step": 6610}, {"loss": 1.0549, "grad_norm": 1.369553565979004, "learning_rate": 0.0002, "epoch": 5.137757081878153, "step": 6620}, {"loss": 1.0056, "grad_norm": 1.177964687347412, "learning_rate": 0.0002, "epoch": 5.145518044237486, "step": 6630}, {"loss": 1.1025, "grad_norm": 1.1397041082382202, "learning_rate": 0.0002, "epoch": 5.153279006596818, "step": 6640}, {"loss": 1.0437, "grad_norm": 1.3976861238479614, "learning_rate": 0.0002, "epoch": 5.161039968956151, "step": 6650}, {"loss": 1.0454, "grad_norm": 1.4824495315551758, "learning_rate": 0.0002, "epoch": 5.1688009313154835, "step": 6660}, {"loss": 1.0356, "grad_norm": 1.2653018236160278, "learning_rate": 0.0002, "epoch": 5.1765618936748155, "step": 6670}, {"loss": 0.9971, "grad_norm": 1.3106069564819336, "learning_rate": 0.0002, "epoch": 5.184322856034148, "step": 6680}, {"loss": 1.0561, "grad_norm": 1.3140279054641724, "learning_rate": 0.0002, "epoch": 5.192083818393481, "step": 6690}, {"loss": 1.0618, "grad_norm": 1.3900256156921387, "learning_rate": 0.0002, "epoch": 5.199844780752813, "step": 6700}, {"loss": 1.0285, "grad_norm": 1.3191124200820923, "learning_rate": 0.0002, "epoch": 5.207605743112146, "step": 6710}, {"loss": 0.9921, "grad_norm": 1.176107406616211, "learning_rate": 0.0002, "epoch": 5.215366705471478, "step": 6720}, {"loss": 1.064, "grad_norm": 1.2364883422851562, "learning_rate": 0.0002, "epoch": 5.223127667830811, "step": 6730}, {"loss": 0.9599, "grad_norm": 1.343022108078003, "learning_rate": 0.0002, "epoch": 5.230888630190144, "step": 6740}, {"loss": 1.0342, "grad_norm": 1.2826898097991943, "learning_rate": 0.0002, "epoch": 5.238649592549476, "step": 6750}, {"loss": 1.0703, "grad_norm": 1.500257134437561, "learning_rate": 0.0002, "epoch": 5.246410554908809, "step": 6760}, {"loss": 1.0114, "grad_norm": 1.2605743408203125, "learning_rate": 0.0002, "epoch": 5.254171517268142, "step": 6770}, {"loss": 1.0825, "grad_norm": 1.2355525493621826, "learning_rate": 0.0002, "epoch": 5.261932479627474, "step": 6780}, {"loss": 1.0436, "grad_norm": 1.2845789194107056, "learning_rate": 0.0002, "epoch": 5.2696934419868064, "step": 6790}, {"loss": 0.989, "grad_norm": 1.3696625232696533, "learning_rate": 0.0002, "epoch": 5.277454404346139, "step": 6800}, {"loss": 1.0991, "grad_norm": 1.4051260948181152, "learning_rate": 0.0002, "epoch": 5.285215366705471, "step": 6810}, {"loss": 1.0987, "grad_norm": 1.266725778579712, "learning_rate": 0.0002, "epoch": 5.292976329064804, "step": 6820}, {"loss": 1.0489, "grad_norm": 1.3475236892700195, "learning_rate": 0.0002, "epoch": 5.300737291424136, "step": 6830}, {"loss": 1.0264, "grad_norm": 1.54409921169281, "learning_rate": 0.0002, "epoch": 5.308498253783469, "step": 6840}, {"loss": 1.033, "grad_norm": 1.2391985654830933, "learning_rate": 0.0002, "epoch": 5.316259216142802, "step": 6850}, {"loss": 1.1058, "grad_norm": 1.2435699701309204, "learning_rate": 0.0002, "epoch": 5.324020178502134, "step": 6860}, {"loss": 1.0179, "grad_norm": 1.8803037405014038, "learning_rate": 0.0002, "epoch": 5.331781140861467, "step": 6870}, {"loss": 0.997, "grad_norm": 1.4195542335510254, "learning_rate": 0.0002, "epoch": 5.3395421032208, "step": 6880}, {"loss": 1.0273, "grad_norm": 1.1853394508361816, "learning_rate": 0.0002, "epoch": 5.347303065580132, "step": 6890}, {"loss": 1.0668, "grad_norm": 1.4016530513763428, "learning_rate": 0.0002, "epoch": 5.355064027939465, "step": 6900}, {"loss": 1.1099, "grad_norm": 1.294339895248413, "learning_rate": 0.0002, "epoch": 5.3628249902987974, "step": 6910}, {"loss": 1.0724, "grad_norm": 1.2952708005905151, "learning_rate": 0.0002, "epoch": 5.370585952658129, "step": 6920}, {"loss": 1.0098, "grad_norm": 1.1361510753631592, "learning_rate": 0.0002, "epoch": 5.378346915017462, "step": 6930}, {"loss": 1.0796, "grad_norm": 1.125805377960205, "learning_rate": 0.0002, "epoch": 5.386107877376794, "step": 6940}, {"loss": 1.122, "grad_norm": 1.1453300714492798, "learning_rate": 0.0002, "epoch": 5.393868839736127, "step": 6950}, {"loss": 1.0977, "grad_norm": 1.4542768001556396, "learning_rate": 0.0002, "epoch": 5.40162980209546, "step": 6960}, {"loss": 1.0825, "grad_norm": 1.2360988855361938, "learning_rate": 0.0002, "epoch": 5.409390764454792, "step": 6970}, {"loss": 1.0631, "grad_norm": 1.2182754278182983, "learning_rate": 0.0002, "epoch": 5.417151726814125, "step": 6980}, {"loss": 1.0471, "grad_norm": 1.2018693685531616, "learning_rate": 0.0002, "epoch": 5.424912689173458, "step": 6990}, {"loss": 1.108, "grad_norm": 1.346124291419983, "learning_rate": 0.0002, "epoch": 5.43267365153279, "step": 7000}, {"loss": 1.0534, "grad_norm": 1.2534189224243164, "learning_rate": 0.0002, "epoch": 5.440434613892123, "step": 7010}, {"loss": 1.0696, "grad_norm": 1.2033339738845825, "learning_rate": 0.0002, "epoch": 5.448195576251456, "step": 7020}, {"loss": 1.0714, "grad_norm": 1.2788134813308716, "learning_rate": 0.0002, "epoch": 5.4559565386107876, "step": 7030}, {"loss": 1.1274, "grad_norm": 1.2751542329788208, "learning_rate": 0.0002, "epoch": 5.46371750097012, "step": 7040}, {"loss": 1.0767, "grad_norm": 1.3237019777297974, "learning_rate": 0.0002, "epoch": 5.471478463329452, "step": 7050}, {"loss": 1.1081, "grad_norm": 1.4932852983474731, "learning_rate": 0.0002, "epoch": 5.479239425688785, "step": 7060}, {"loss": 1.0197, "grad_norm": 1.4003876447677612, "learning_rate": 0.0002, "epoch": 5.487000388048118, "step": 7070}, {"loss": 1.0662, "grad_norm": 1.404799461364746, "learning_rate": 0.0002, "epoch": 5.49476135040745, "step": 7080}, {"loss": 1.0354, "grad_norm": 1.4486982822418213, "learning_rate": 0.0002, "epoch": 5.502522312766783, "step": 7090}, {"loss": 1.0645, "grad_norm": 1.1713480949401855, "learning_rate": 0.0002, "epoch": 5.510283275126116, "step": 7100}, {"loss": 1.006, "grad_norm": 1.4062601327896118, "learning_rate": 0.0002, "epoch": 5.518044237485448, "step": 7110}, {"loss": 1.0459, "grad_norm": 1.211629867553711, "learning_rate": 0.0002, "epoch": 5.525805199844781, "step": 7120}, {"loss": 1.102, "grad_norm": 1.2523176670074463, "learning_rate": 0.0002, "epoch": 5.533566162204114, "step": 7130}, {"loss": 1.1132, "grad_norm": 1.4467198848724365, "learning_rate": 0.0002, "epoch": 5.541327124563446, "step": 7140}, {"loss": 1.1557, "grad_norm": 1.5961614847183228, "learning_rate": 0.0002, "epoch": 5.5490880869227786, "step": 7150}, {"loss": 1.0859, "grad_norm": 1.320656418800354, "learning_rate": 0.0002, "epoch": 5.5568490492821105, "step": 7160}, {"loss": 1.109, "grad_norm": 1.2423332929611206, "learning_rate": 0.0002, "epoch": 5.564610011641443, "step": 7170}, {"loss": 1.0046, "grad_norm": 1.2919669151306152, "learning_rate": 0.0002, "epoch": 5.572370974000776, "step": 7180}, {"loss": 1.046, "grad_norm": 1.1678385734558105, "learning_rate": 0.0002, "epoch": 5.580131936360108, "step": 7190}, {"loss": 1.1011, "grad_norm": 1.4250764846801758, "learning_rate": 0.0002, "epoch": 5.587892898719441, "step": 7200}, {"loss": 1.1254, "grad_norm": 1.5308716297149658, "learning_rate": 0.0002, "epoch": 5.595653861078774, "step": 7210}, {"loss": 1.121, "grad_norm": 1.2678815126419067, "learning_rate": 0.0002, "epoch": 5.603414823438106, "step": 7220}, {"loss": 1.0846, "grad_norm": 1.127856969833374, "learning_rate": 0.0002, "epoch": 5.611175785797439, "step": 7230}, {"loss": 1.0647, "grad_norm": 1.3832560777664185, "learning_rate": 0.0002, "epoch": 5.618936748156772, "step": 7240}, {"loss": 1.0658, "grad_norm": 1.3226919174194336, "learning_rate": 0.0002, "epoch": 5.626697710516104, "step": 7250}, {"loss": 1.1175, "grad_norm": 1.3418006896972656, "learning_rate": 0.0002, "epoch": 5.634458672875437, "step": 7260}, {"loss": 1.0956, "grad_norm": 1.2625300884246826, "learning_rate": 0.0002, "epoch": 5.642219635234769, "step": 7270}, {"loss": 1.067, "grad_norm": 1.1579464673995972, "learning_rate": 0.0002, "epoch": 5.6499805975941015, "step": 7280}, {"loss": 1.0447, "grad_norm": 1.4998650550842285, "learning_rate": 0.0002, "epoch": 5.657741559953434, "step": 7290}, {"loss": 1.1256, "grad_norm": 1.2670758962631226, "learning_rate": 0.0002, "epoch": 5.665502522312766, "step": 7300}, {"loss": 1.1267, "grad_norm": 1.2959760427474976, "learning_rate": 0.0002, "epoch": 5.673263484672099, "step": 7310}, {"loss": 1.1387, "grad_norm": 1.2460671663284302, "learning_rate": 0.0002, "epoch": 5.681024447031432, "step": 7320}, {"loss": 1.0756, "grad_norm": 1.1313989162445068, "learning_rate": 0.0002, "epoch": 5.688785409390764, "step": 7330}, {"loss": 1.0618, "grad_norm": 1.282527208328247, "learning_rate": 0.0002, "epoch": 5.696546371750097, "step": 7340}, {"loss": 1.1315, "grad_norm": 1.3380206823349, "learning_rate": 0.0002, "epoch": 5.70430733410943, "step": 7350}, {"loss": 1.0949, "grad_norm": 1.1648279428482056, "learning_rate": 0.0002, "epoch": 5.712068296468762, "step": 7360}, {"loss": 1.1705, "grad_norm": 1.3059816360473633, "learning_rate": 0.0002, "epoch": 5.719829258828095, "step": 7370}, {"loss": 1.1496, "grad_norm": 1.1905046701431274, "learning_rate": 0.0002, "epoch": 5.727590221187427, "step": 7380}, {"loss": 1.1356, "grad_norm": 1.4089630842208862, "learning_rate": 0.0002, "epoch": 5.73535118354676, "step": 7390}, {"loss": 1.1349, "grad_norm": 1.256721019744873, "learning_rate": 0.0002, "epoch": 5.7431121459060925, "step": 7400}, {"loss": 1.0682, "grad_norm": 1.1915162801742554, "learning_rate": 0.0002, "epoch": 5.7508731082654245, "step": 7410}, {"loss": 1.1257, "grad_norm": 1.1935480833053589, "learning_rate": 0.0002, "epoch": 5.758634070624757, "step": 7420}, {"loss": 1.1348, "grad_norm": 1.1761008501052856, "learning_rate": 0.0002, "epoch": 5.76639503298409, "step": 7430}, {"loss": 1.0837, "grad_norm": 1.2540549039840698, "learning_rate": 0.0002, "epoch": 5.774155995343422, "step": 7440}, {"loss": 1.1527, "grad_norm": 1.5295120477676392, "learning_rate": 0.0002, "epoch": 5.781916957702755, "step": 7450}, {"loss": 1.1146, "grad_norm": 1.1081160306930542, "learning_rate": 0.0002, "epoch": 5.789677920062088, "step": 7460}, {"loss": 1.1304, "grad_norm": 1.4381253719329834, "learning_rate": 0.0002, "epoch": 5.79743888242142, "step": 7470}, {"loss": 1.0684, "grad_norm": 1.3079341650009155, "learning_rate": 0.0002, "epoch": 5.805199844780753, "step": 7480}, {"loss": 1.0544, "grad_norm": 1.1372792720794678, "learning_rate": 0.0002, "epoch": 5.812960807140085, "step": 7490}, {"loss": 1.1622, "grad_norm": 1.3221744298934937, "learning_rate": 0.0002, "epoch": 5.820721769499418, "step": 7500}, {"loss": 1.1515, "grad_norm": 1.3436939716339111, "learning_rate": 0.0002, "epoch": 5.828482731858751, "step": 7510}, {"loss": 1.1154, "grad_norm": 1.3916879892349243, "learning_rate": 0.0002, "epoch": 5.8362436942180835, "step": 7520}, {"loss": 1.0816, "grad_norm": 1.2463704347610474, "learning_rate": 0.0002, "epoch": 5.8440046565774155, "step": 7530}, {"loss": 1.0745, "grad_norm": 1.097051739692688, "learning_rate": 0.0002, "epoch": 5.851765618936748, "step": 7540}, {"loss": 1.1454, "grad_norm": 1.1554739475250244, "learning_rate": 0.0002, "epoch": 5.85952658129608, "step": 7550}, {"loss": 1.0953, "grad_norm": 1.2384694814682007, "learning_rate": 0.0002, "epoch": 5.867287543655413, "step": 7560}, {"loss": 1.1734, "grad_norm": 1.142815351486206, "learning_rate": 0.0002, "epoch": 5.875048506014746, "step": 7570}, {"loss": 1.162, "grad_norm": 1.3637062311172485, "learning_rate": 0.0002, "epoch": 5.882809468374078, "step": 7580}, {"loss": 1.0781, "grad_norm": 1.2449073791503906, "learning_rate": 0.0002, "epoch": 5.890570430733411, "step": 7590}, {"loss": 1.1191, "grad_norm": 1.358058214187622, "learning_rate": 0.0002, "epoch": 5.898331393092743, "step": 7600}, {"loss": 1.0779, "grad_norm": 1.264655351638794, "learning_rate": 0.0002, "epoch": 5.906092355452076, "step": 7610}, {"loss": 1.1538, "grad_norm": 1.3186019659042358, "learning_rate": 0.0002, "epoch": 5.913853317811409, "step": 7620}, {"loss": 1.1076, "grad_norm": 1.4111460447311401, "learning_rate": 0.0002, "epoch": 5.921614280170742, "step": 7630}, {"loss": 1.1765, "grad_norm": 1.1078972816467285, "learning_rate": 0.0002, "epoch": 5.929375242530074, "step": 7640}, {"loss": 1.1305, "grad_norm": 1.2742213010787964, "learning_rate": 0.0002, "epoch": 5.9371362048894065, "step": 7650}, {"loss": 1.144, "grad_norm": 1.3412781953811646, "learning_rate": 0.0002, "epoch": 5.9448971672487385, "step": 7660}, {"loss": 1.1642, "grad_norm": 1.123005986213684, "learning_rate": 0.0002, "epoch": 5.952658129608071, "step": 7670}, {"loss": 1.0732, "grad_norm": 1.2203444242477417, "learning_rate": 0.0002, "epoch": 5.960419091967404, "step": 7680}, {"loss": 1.158, "grad_norm": 1.341011643409729, "learning_rate": 0.0002, "epoch": 5.968180054326736, "step": 7690}, {"loss": 1.1144, "grad_norm": 1.2689454555511475, "learning_rate": 0.0002, "epoch": 5.975941016686069, "step": 7700}, {"loss": 1.2051, "grad_norm": 1.1518112421035767, "learning_rate": 0.0002, "epoch": 5.983701979045401, "step": 7710}, {"loss": 1.1868, "grad_norm": 1.3698320388793945, "learning_rate": 0.0002, "epoch": 5.991462941404734, "step": 7720}, {"loss": 1.0651, "grad_norm": 1.2812788486480713, "learning_rate": 0.0002, "epoch": 5.999223903764067, "step": 7730}, {"eval_loss": 2.252762794494629, "eval_runtime": 114.8471, "eval_samples_per_second": 4.415, "eval_steps_per_second": 0.557, "epoch": 6.0, "step": 7731}, {"loss": 0.8629, "grad_norm": 1.8642009496688843, "learning_rate": 0.0002, "epoch": 6.006984866123399, "step": 7740}, {"loss": 0.8435, "grad_norm": 1.7081232070922852, "learning_rate": 0.0002, "epoch": 6.014745828482732, "step": 7750}, {"loss": 0.7729, "grad_norm": 1.6233899593353271, "learning_rate": 0.0002, "epoch": 6.022506790842065, "step": 7760}, {"loss": 0.7907, "grad_norm": 1.5111888647079468, "learning_rate": 0.0002, "epoch": 6.030267753201397, "step": 7770}, {"loss": 0.7908, "grad_norm": 1.5278418064117432, "learning_rate": 0.0002, "epoch": 6.0380287155607295, "step": 7780}, {"loss": 0.835, "grad_norm": 1.5932185649871826, "learning_rate": 0.0002, "epoch": 6.045789677920062, "step": 7790}, {"loss": 0.7682, "grad_norm": 1.5990597009658813, "learning_rate": 0.0002, "epoch": 6.053550640279394, "step": 7800}, {"loss": 0.8559, "grad_norm": 1.7498669624328613, "learning_rate": 0.0002, "epoch": 6.061311602638727, "step": 7810}, {"loss": 0.8069, "grad_norm": 1.6105555295944214, "learning_rate": 0.0002, "epoch": 6.06907256499806, "step": 7820}, {"loss": 0.8473, "grad_norm": 1.5214293003082275, "learning_rate": 0.0002, "epoch": 6.076833527357392, "step": 7830}, {"loss": 0.8328, "grad_norm": 1.6586973667144775, "learning_rate": 0.0002, "epoch": 6.084594489716725, "step": 7840}, {"loss": 0.8415, "grad_norm": 1.467391848564148, "learning_rate": 0.0002, "epoch": 6.092355452076057, "step": 7850}, {"loss": 0.8274, "grad_norm": 1.537361741065979, "learning_rate": 0.0002, "epoch": 6.10011641443539, "step": 7860}, {"loss": 0.8011, "grad_norm": 1.621764898300171, "learning_rate": 0.0002, "epoch": 6.107877376794723, "step": 7870}, {"loss": 0.8556, "grad_norm": 1.583751916885376, "learning_rate": 0.0002, "epoch": 6.115638339154055, "step": 7880}, {"loss": 0.8829, "grad_norm": 1.6199619770050049, "learning_rate": 0.0002, "epoch": 6.123399301513388, "step": 7890}, {"loss": 0.8226, "grad_norm": 1.6163095235824585, "learning_rate": 0.0002, "epoch": 6.1311602638727205, "step": 7900}, {"loss": 0.8203, "grad_norm": 1.6120976209640503, "learning_rate": 0.0002, "epoch": 6.1389212262320525, "step": 7910}, {"loss": 0.7915, "grad_norm": 1.7886850833892822, "learning_rate": 0.0002, "epoch": 6.146682188591385, "step": 7920}, {"loss": 0.7808, "grad_norm": 1.408303141593933, "learning_rate": 0.0002, "epoch": 6.154443150950718, "step": 7930}, {"loss": 0.8404, "grad_norm": 1.6048113107681274, "learning_rate": 0.0002, "epoch": 6.16220411331005, "step": 7940}, {"loss": 0.8705, "grad_norm": 1.424306869506836, "learning_rate": 0.0002, "epoch": 6.169965075669383, "step": 7950}, {"loss": 0.8177, "grad_norm": 1.4453672170639038, "learning_rate": 0.0002, "epoch": 6.177726038028716, "step": 7960}, {"loss": 0.8182, "grad_norm": 1.3157061338424683, "learning_rate": 0.0002, "epoch": 6.185487000388048, "step": 7970}, {"loss": 0.891, "grad_norm": 1.330541729927063, "learning_rate": 0.0002, "epoch": 6.193247962747381, "step": 7980}, {"loss": 0.8599, "grad_norm": 1.6306229829788208, "learning_rate": 0.0002, "epoch": 6.201008925106713, "step": 7990}, {"loss": 0.9069, "grad_norm": 1.6332136392593384, "learning_rate": 0.0002, "epoch": 6.208769887466046, "step": 8000}, {"loss": 0.83, "grad_norm": 1.708613395690918, "learning_rate": 0.0002, "epoch": 6.216530849825379, "step": 8010}, {"loss": 0.8509, "grad_norm": 1.6637346744537354, "learning_rate": 0.0002, "epoch": 6.224291812184711, "step": 8020}, {"loss": 0.84, "grad_norm": 1.5675315856933594, "learning_rate": 0.0002, "epoch": 6.2320527745440435, "step": 8030}, {"loss": 0.8491, "grad_norm": 1.5826327800750732, "learning_rate": 0.0002, "epoch": 6.239813736903376, "step": 8040}, {"loss": 0.8374, "grad_norm": 1.7382984161376953, "learning_rate": 0.0002, "epoch": 6.247574699262708, "step": 8050}, {"loss": 0.8795, "grad_norm": 1.5272295475006104, "learning_rate": 0.0002, "epoch": 6.255335661622041, "step": 8060}, {"loss": 0.8745, "grad_norm": 1.8195022344589233, "learning_rate": 0.0002, "epoch": 6.263096623981374, "step": 8070}, {"loss": 0.8743, "grad_norm": 1.679901361465454, "learning_rate": 0.0002, "epoch": 6.270857586340706, "step": 8080}, {"loss": 0.9006, "grad_norm": 1.4921348094940186, "learning_rate": 0.0002, "epoch": 6.278618548700039, "step": 8090}, {"loss": 0.899, "grad_norm": 1.4627857208251953, "learning_rate": 0.0002, "epoch": 6.286379511059371, "step": 8100}, {"loss": 0.8944, "grad_norm": 1.3528631925582886, "learning_rate": 0.0002, "epoch": 6.294140473418704, "step": 8110}, {"loss": 0.9355, "grad_norm": 1.6863102912902832, "learning_rate": 0.0002, "epoch": 6.301901435778037, "step": 8120}, {"loss": 0.8764, "grad_norm": 1.6178052425384521, "learning_rate": 0.0002, "epoch": 6.309662398137369, "step": 8130}, {"loss": 0.9182, "grad_norm": 1.7626280784606934, "learning_rate": 0.0002, "epoch": 6.317423360496702, "step": 8140}, {"loss": 0.8886, "grad_norm": 1.7188845872879028, "learning_rate": 0.0002, "epoch": 6.3251843228560345, "step": 8150}, {"loss": 0.895, "grad_norm": 1.5777133703231812, "learning_rate": 0.0002, "epoch": 6.3329452852153665, "step": 8160}, {"loss": 0.9247, "grad_norm": 1.7653207778930664, "learning_rate": 0.0002, "epoch": 6.340706247574699, "step": 8170}, {"loss": 0.8003, "grad_norm": 1.6861237287521362, "learning_rate": 0.0002, "epoch": 6.348467209934032, "step": 8180}, {"loss": 0.884, "grad_norm": 1.6318124532699585, "learning_rate": 0.0002, "epoch": 6.356228172293364, "step": 8190}, {"loss": 0.8341, "grad_norm": 1.6192939281463623, "learning_rate": 0.0002, "epoch": 6.363989134652697, "step": 8200}, {"loss": 0.8939, "grad_norm": 1.7641773223876953, "learning_rate": 0.0002, "epoch": 6.371750097012029, "step": 8210}, {"loss": 0.8582, "grad_norm": 1.6470493078231812, "learning_rate": 0.0002, "epoch": 6.379511059371362, "step": 8220}, {"loss": 0.9351, "grad_norm": 1.5898468494415283, "learning_rate": 0.0002, "epoch": 6.387272021730695, "step": 8230}, {"loss": 0.9658, "grad_norm": 1.8025981187820435, "learning_rate": 0.0002, "epoch": 6.395032984090027, "step": 8240}, {"loss": 0.8953, "grad_norm": 1.7035106420516968, "learning_rate": 0.0002, "epoch": 6.40279394644936, "step": 8250}, {"loss": 0.9193, "grad_norm": 1.5968799591064453, "learning_rate": 0.0002, "epoch": 6.410554908808693, "step": 8260}, {"loss": 0.929, "grad_norm": 1.7492800951004028, "learning_rate": 0.0002, "epoch": 6.418315871168025, "step": 8270}, {"loss": 0.9297, "grad_norm": 1.6914138793945312, "learning_rate": 0.0002, "epoch": 6.4260768335273575, "step": 8280}, {"loss": 0.8878, "grad_norm": 1.5761380195617676, "learning_rate": 0.0002, "epoch": 6.43383779588669, "step": 8290}, {"loss": 0.8761, "grad_norm": 1.5164411067962646, "learning_rate": 0.0002, "epoch": 6.441598758246022, "step": 8300}, {"loss": 0.88, "grad_norm": 1.6600215435028076, "learning_rate": 0.0002, "epoch": 6.449359720605355, "step": 8310}, {"loss": 0.9113, "grad_norm": 1.2477679252624512, "learning_rate": 0.0002, "epoch": 6.457120682964687, "step": 8320}, {"loss": 0.8822, "grad_norm": 1.3698599338531494, "learning_rate": 0.0002, "epoch": 6.46488164532402, "step": 8330}, {"loss": 0.9295, "grad_norm": 1.4847341775894165, "learning_rate": 0.0002, "epoch": 6.472642607683353, "step": 8340}, {"loss": 0.9243, "grad_norm": 1.4713412523269653, "learning_rate": 0.0002, "epoch": 6.480403570042685, "step": 8350}, {"loss": 0.9102, "grad_norm": 1.334523320198059, "learning_rate": 0.0002, "epoch": 6.488164532402018, "step": 8360}, {"loss": 0.8563, "grad_norm": 2.0054359436035156, "learning_rate": 0.0002, "epoch": 6.495925494761351, "step": 8370}, {"loss": 0.9759, "grad_norm": 1.560014247894287, "learning_rate": 0.0002, "epoch": 6.503686457120683, "step": 8380}, {"loss": 0.8542, "grad_norm": 1.518526554107666, "learning_rate": 0.0002, "epoch": 6.511447419480016, "step": 8390}, {"loss": 0.937, "grad_norm": 1.3841272592544556, "learning_rate": 0.0002, "epoch": 6.5192083818393485, "step": 8400}, {"loss": 0.9576, "grad_norm": 1.5191527605056763, "learning_rate": 0.0002, "epoch": 6.5269693441986805, "step": 8410}, {"loss": 0.8899, "grad_norm": 1.5275579690933228, "learning_rate": 0.0002, "epoch": 6.534730306558013, "step": 8420}, {"loss": 0.9291, "grad_norm": 1.621590256690979, "learning_rate": 0.0002, "epoch": 6.542491268917345, "step": 8430}, {"loss": 0.9011, "grad_norm": 1.7939082384109497, "learning_rate": 0.0002, "epoch": 6.550252231276678, "step": 8440}, {"loss": 0.8896, "grad_norm": 1.4542964696884155, "learning_rate": 0.0002, "epoch": 6.558013193636011, "step": 8450}, {"loss": 0.9393, "grad_norm": 1.5458455085754395, "learning_rate": 0.0002, "epoch": 6.565774155995343, "step": 8460}, {"loss": 0.9028, "grad_norm": 1.550359845161438, "learning_rate": 0.0002, "epoch": 6.573535118354676, "step": 8470}, {"loss": 0.9271, "grad_norm": 1.527757167816162, "learning_rate": 0.0002, "epoch": 6.581296080714009, "step": 8480}, {"loss": 0.966, "grad_norm": 1.4683486223220825, "learning_rate": 0.0002, "epoch": 6.589057043073341, "step": 8490}, {"loss": 0.9079, "grad_norm": 1.5057084560394287, "learning_rate": 0.0002, "epoch": 6.596818005432674, "step": 8500}, {"loss": 0.9235, "grad_norm": 1.648289442062378, "learning_rate": 0.0002, "epoch": 6.604578967792007, "step": 8510}, {"loss": 0.9113, "grad_norm": 1.578914761543274, "learning_rate": 0.0002, "epoch": 6.612339930151339, "step": 8520}, {"loss": 0.8894, "grad_norm": 1.5064080953598022, "learning_rate": 0.0002, "epoch": 6.6201008925106715, "step": 8530}, {"loss": 0.8981, "grad_norm": 1.5717744827270508, "learning_rate": 0.0002, "epoch": 6.6278618548700035, "step": 8540}, {"loss": 0.887, "grad_norm": 1.7954767942428589, "learning_rate": 0.0002, "epoch": 6.635622817229336, "step": 8550}, {"loss": 0.927, "grad_norm": 1.6172343492507935, "learning_rate": 0.0002, "epoch": 6.643383779588669, "step": 8560}, {"loss": 0.9384, "grad_norm": 1.6627886295318604, "learning_rate": 0.0002, "epoch": 6.651144741948001, "step": 8570}, {"loss": 0.959, "grad_norm": 1.5264919996261597, "learning_rate": 0.0002, "epoch": 6.658905704307334, "step": 8580}, {"loss": 0.9103, "grad_norm": 1.609248161315918, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 8590}, {"loss": 0.9395, "grad_norm": 1.5474581718444824, "learning_rate": 0.0002, "epoch": 6.674427629025999, "step": 8600}, {"loss": 0.9018, "grad_norm": 1.6294898986816406, "learning_rate": 0.0002, "epoch": 6.682188591385332, "step": 8610}, {"loss": 0.9323, "grad_norm": 1.612615942955017, "learning_rate": 0.0002, "epoch": 6.689949553744665, "step": 8620}, {"loss": 0.9218, "grad_norm": 1.741325855255127, "learning_rate": 0.0002, "epoch": 6.697710516103997, "step": 8630}, {"loss": 1.0475, "grad_norm": 1.5089004039764404, "learning_rate": 0.0002, "epoch": 6.70547147846333, "step": 8640}, {"loss": 1.0009, "grad_norm": 1.4725582599639893, "learning_rate": 0.0002, "epoch": 6.713232440822662, "step": 8650}, {"loss": 0.9818, "grad_norm": 1.6992095708847046, "learning_rate": 0.0002, "epoch": 6.7209934031819945, "step": 8660}, {"loss": 0.9229, "grad_norm": 1.5938470363616943, "learning_rate": 0.0002, "epoch": 6.728754365541327, "step": 8670}, {"loss": 0.9411, "grad_norm": 1.58723783493042, "learning_rate": 0.0002, "epoch": 6.736515327900659, "step": 8680}, {"loss": 0.9738, "grad_norm": 1.514389991760254, "learning_rate": 0.0002, "epoch": 6.744276290259992, "step": 8690}, {"loss": 0.9283, "grad_norm": 1.6799157857894897, "learning_rate": 0.0002, "epoch": 6.752037252619325, "step": 8700}, {"loss": 0.9138, "grad_norm": 1.5436359643936157, "learning_rate": 0.0002, "epoch": 6.759798214978657, "step": 8710}, {"loss": 0.9678, "grad_norm": 1.477137565612793, "learning_rate": 0.0002, "epoch": 6.76755917733799, "step": 8720}, {"loss": 1.0044, "grad_norm": 1.7383503913879395, "learning_rate": 0.0002, "epoch": 6.775320139697323, "step": 8730}, {"loss": 0.9492, "grad_norm": 1.8000324964523315, "learning_rate": 0.0002, "epoch": 6.783081102056655, "step": 8740}, {"loss": 0.8943, "grad_norm": 1.3099453449249268, "learning_rate": 0.0002, "epoch": 6.790842064415988, "step": 8750}, {"loss": 0.9709, "grad_norm": 1.8775172233581543, "learning_rate": 0.0002, "epoch": 6.79860302677532, "step": 8760}, {"loss": 0.9356, "grad_norm": 1.5832085609436035, "learning_rate": 0.0002, "epoch": 6.806363989134653, "step": 8770}, {"loss": 0.9397, "grad_norm": 1.4903252124786377, "learning_rate": 0.0002, "epoch": 6.8141249514939854, "step": 8780}, {"loss": 0.9602, "grad_norm": 1.6360470056533813, "learning_rate": 0.0002, "epoch": 6.821885913853317, "step": 8790}, {"loss": 0.957, "grad_norm": 1.5457707643508911, "learning_rate": 0.0002, "epoch": 6.82964687621265, "step": 8800}, {"loss": 0.943, "grad_norm": 1.5449066162109375, "learning_rate": 0.0002, "epoch": 6.837407838571983, "step": 8810}, {"loss": 1.0007, "grad_norm": 1.618337631225586, "learning_rate": 0.0002, "epoch": 6.845168800931315, "step": 8820}, {"loss": 0.9314, "grad_norm": 1.38296639919281, "learning_rate": 0.0002, "epoch": 6.852929763290648, "step": 8830}, {"loss": 0.9349, "grad_norm": 1.6427991390228271, "learning_rate": 0.0002, "epoch": 6.860690725649981, "step": 8840}, {"loss": 1.0194, "grad_norm": 1.4980270862579346, "learning_rate": 0.0002, "epoch": 6.868451688009313, "step": 8850}, {"loss": 0.9541, "grad_norm": 1.3800020217895508, "learning_rate": 0.0002, "epoch": 6.876212650368646, "step": 8860}, {"loss": 1.0102, "grad_norm": 1.5971838235855103, "learning_rate": 0.0002, "epoch": 6.883973612727978, "step": 8870}, {"loss": 1.0105, "grad_norm": 1.4429489374160767, "learning_rate": 0.0002, "epoch": 6.891734575087311, "step": 8880}, {"loss": 0.9143, "grad_norm": 1.4959166049957275, "learning_rate": 0.0002, "epoch": 6.899495537446644, "step": 8890}, {"loss": 0.9403, "grad_norm": 1.5776222944259644, "learning_rate": 0.0002, "epoch": 6.907256499805976, "step": 8900}, {"loss": 0.9256, "grad_norm": 1.510412573814392, "learning_rate": 0.0002, "epoch": 6.915017462165308, "step": 8910}, {"loss": 1.0095, "grad_norm": 1.7216295003890991, "learning_rate": 0.0002, "epoch": 6.922778424524641, "step": 8920}, {"loss": 0.9464, "grad_norm": 1.830762505531311, "learning_rate": 0.0002, "epoch": 6.930539386883973, "step": 8930}, {"loss": 0.9704, "grad_norm": 1.3472434282302856, "learning_rate": 0.0002, "epoch": 6.938300349243306, "step": 8940}, {"loss": 0.9718, "grad_norm": 1.5748040676116943, "learning_rate": 0.0002, "epoch": 6.946061311602639, "step": 8950}, {"loss": 0.9891, "grad_norm": 1.5317506790161133, "learning_rate": 0.0002, "epoch": 6.953822273961971, "step": 8960}, {"loss": 0.9513, "grad_norm": 1.5565721988677979, "learning_rate": 0.0002, "epoch": 6.961583236321304, "step": 8970}, {"loss": 0.9118, "grad_norm": 1.5288970470428467, "learning_rate": 0.0002, "epoch": 6.969344198680636, "step": 8980}, {"loss": 0.9789, "grad_norm": 1.562624454498291, "learning_rate": 0.0002, "epoch": 6.977105161039969, "step": 8990}, {"loss": 0.9929, "grad_norm": 1.3777633905410767, "learning_rate": 0.0002, "epoch": 6.984866123399302, "step": 9000}, {"loss": 0.9713, "grad_norm": 1.5868972539901733, "learning_rate": 0.0002, "epoch": 6.992627085758635, "step": 9010}]}
+{"epoch": 7.996895615056267, "step": 10304, "epoch_duration": 3950.4752156734467, "total_accumulated_duration": 29862.658538103104, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 28774.0}, "peak_memory_reserved": {"GPU_0": 28774.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-10000/checkpoint-2577", "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": 3.0855, "grad_norm": 1.0751162767410278, "learning_rate": 0.0002, "epoch": 0.007760962359332557, "step": 10}, {"loss": 2.4744, "grad_norm": 0.4697345793247223, "learning_rate": 0.0002, "epoch": 0.015521924718665115, "step": 20}, {"loss": 2.193, "grad_norm": 0.5370839238166809, "learning_rate": 0.0002, "epoch": 0.023282887077997673, "step": 30}, {"loss": 2.0599, "grad_norm": 0.46794816851615906, "learning_rate": 0.0002, "epoch": 0.03104384943733023, "step": 40}, {"loss": 1.9354, "grad_norm": 0.44624820351600647, "learning_rate": 0.0002, "epoch": 0.038804811796662786, "step": 50}, {"loss": 1.9319, "grad_norm": 0.3953201472759247, "learning_rate": 0.0002, "epoch": 0.046565774155995346, "step": 60}, {"loss": 1.9099, "grad_norm": 0.3935912549495697, "learning_rate": 0.0002, "epoch": 0.0543267365153279, "step": 70}, {"loss": 1.8795, "grad_norm": 0.4520699381828308, "learning_rate": 0.0002, "epoch": 0.06208769887466046, "step": 80}, {"loss": 1.8354, "grad_norm": 0.3801847994327545, "learning_rate": 0.0002, "epoch": 0.06984866123399301, "step": 90}, {"loss": 1.9053, "grad_norm": 0.4020165205001831, "learning_rate": 0.0002, "epoch": 0.07760962359332557, "step": 100}, {"loss": 1.8779, "grad_norm": 0.3860672116279602, "learning_rate": 0.0002, "epoch": 0.08537058595265813, "step": 110}, {"loss": 1.8731, "grad_norm": 0.3681113123893738, "learning_rate": 0.0002, "epoch": 0.09313154831199069, "step": 120}, {"loss": 1.8157, "grad_norm": 0.3594866991043091, "learning_rate": 0.0002, "epoch": 0.10089251067132324, "step": 130}, {"loss": 1.8266, "grad_norm": 0.3879193663597107, "learning_rate": 0.0002, "epoch": 0.1086534730306558, "step": 140}, {"loss": 1.8818, "grad_norm": 0.3270505666732788, "learning_rate": 0.0002, "epoch": 0.11641443538998836, "step": 150}, {"loss": 1.87, "grad_norm": 0.36824458837509155, "learning_rate": 0.0002, "epoch": 0.12417539774932092, "step": 160}, {"loss": 1.8305, "grad_norm": 0.383882075548172, "learning_rate": 0.0002, "epoch": 0.13193636010865348, "step": 170}, {"loss": 1.8584, "grad_norm": 0.3368665874004364, "learning_rate": 0.0002, "epoch": 0.13969732246798602, "step": 180}, {"loss": 1.7882, "grad_norm": 0.35961097478866577, "learning_rate": 0.0002, "epoch": 0.1474582848273186, "step": 190}, {"loss": 1.8467, "grad_norm": 0.3415963351726532, "learning_rate": 0.0002, "epoch": 0.15521924718665114, "step": 200}, {"loss": 1.8543, "grad_norm": 0.4100632071495056, "learning_rate": 0.0002, "epoch": 0.1629802095459837, "step": 210}, {"loss": 1.8226, "grad_norm": 0.3516307473182678, "learning_rate": 0.0002, "epoch": 0.17074117190531626, "step": 220}, {"loss": 1.7386, "grad_norm": 0.37919050455093384, "learning_rate": 0.0002, "epoch": 0.1785021342646488, "step": 230}, {"loss": 1.7937, "grad_norm": 0.33270683884620667, "learning_rate": 0.0002, "epoch": 0.18626309662398138, "step": 240}, {"loss": 1.7925, "grad_norm": 0.3348783254623413, "learning_rate": 0.0002, "epoch": 0.19402405898331393, "step": 250}, {"loss": 1.7774, "grad_norm": 0.3888475298881531, "learning_rate": 0.0002, "epoch": 0.20178502134264648, "step": 260}, {"loss": 1.8381, "grad_norm": 0.3554602861404419, "learning_rate": 0.0002, "epoch": 0.20954598370197905, "step": 270}, {"loss": 1.8359, "grad_norm": 0.33277708292007446, "learning_rate": 0.0002, "epoch": 0.2173069460613116, "step": 280}, {"loss": 1.7713, "grad_norm": 0.3281584680080414, "learning_rate": 0.0002, "epoch": 0.22506790842064417, "step": 290}, {"loss": 1.8181, "grad_norm": 0.3185969591140747, "learning_rate": 0.0002, "epoch": 0.23282887077997672, "step": 300}, {"loss": 1.8595, "grad_norm": 0.35335442423820496, "learning_rate": 0.0002, "epoch": 0.24058983313930926, "step": 310}, {"loss": 1.87, "grad_norm": 0.3119595944881439, "learning_rate": 0.0002, "epoch": 0.24835079549864184, "step": 320}, {"loss": 1.8357, "grad_norm": 0.36424458026885986, "learning_rate": 0.0002, "epoch": 0.2561117578579744, "step": 330}, {"loss": 1.8003, "grad_norm": 0.3618951141834259, "learning_rate": 0.0002, "epoch": 0.26387272021730696, "step": 340}, {"loss": 1.8221, "grad_norm": 0.312757670879364, "learning_rate": 0.0002, "epoch": 0.2716336825766395, "step": 350}, {"loss": 1.9031, "grad_norm": 0.326016366481781, "learning_rate": 0.0002, "epoch": 0.27939464493597205, "step": 360}, {"loss": 1.8214, "grad_norm": 0.34093883633613586, "learning_rate": 0.0002, "epoch": 0.2871556072953046, "step": 370}, {"loss": 1.7733, "grad_norm": 0.32325029373168945, "learning_rate": 0.0002, "epoch": 0.2949165696546372, "step": 380}, {"loss": 1.842, "grad_norm": 0.34105437994003296, "learning_rate": 0.0002, "epoch": 0.30267753201396974, "step": 390}, {"loss": 1.7926, "grad_norm": 0.32565295696258545, "learning_rate": 0.0002, "epoch": 0.3104384943733023, "step": 400}, {"loss": 1.8031, "grad_norm": 0.32742050290107727, "learning_rate": 0.0002, "epoch": 0.31819945673263483, "step": 410}, {"loss": 1.907, "grad_norm": 0.30233046412467957, "learning_rate": 0.0002, "epoch": 0.3259604190919674, "step": 420}, {"loss": 1.7623, "grad_norm": 0.32419222593307495, "learning_rate": 0.0002, "epoch": 0.3337213814513, "step": 430}, {"loss": 1.865, "grad_norm": 0.3653007745742798, "learning_rate": 0.0002, "epoch": 0.3414823438106325, "step": 440}, {"loss": 1.8044, "grad_norm": 0.31617099046707153, "learning_rate": 0.0002, "epoch": 0.3492433061699651, "step": 450}, {"loss": 1.7677, "grad_norm": 0.3305962085723877, "learning_rate": 0.0002, "epoch": 0.3570042685292976, "step": 460}, {"loss": 1.8155, "grad_norm": 0.3178933262825012, "learning_rate": 0.0002, "epoch": 0.36476523088863017, "step": 470}, {"loss": 1.7485, "grad_norm": 0.37163782119750977, "learning_rate": 0.0002, "epoch": 0.37252619324796277, "step": 480}, {"loss": 1.8804, "grad_norm": 0.469844788312912, "learning_rate": 0.0002, "epoch": 0.3802871556072953, "step": 490}, {"loss": 1.8343, "grad_norm": 0.3409338593482971, "learning_rate": 0.0002, "epoch": 0.38804811796662786, "step": 500}, {"loss": 1.8433, "grad_norm": 0.31943467259407043, "learning_rate": 0.0002, "epoch": 0.3958090803259604, "step": 510}, {"loss": 1.7873, "grad_norm": 0.32293614745140076, "learning_rate": 0.0002, "epoch": 0.40357004268529295, "step": 520}, {"loss": 1.8584, "grad_norm": 0.2994382977485657, "learning_rate": 0.0002, "epoch": 0.41133100504462555, "step": 530}, {"loss": 1.8153, "grad_norm": 0.3273141384124756, "learning_rate": 0.0002, "epoch": 0.4190919674039581, "step": 540}, {"loss": 1.8097, "grad_norm": 0.3020550012588501, "learning_rate": 0.0002, "epoch": 0.42685292976329064, "step": 550}, {"loss": 1.8551, "grad_norm": 0.30113112926483154, "learning_rate": 0.0002, "epoch": 0.4346138921226232, "step": 560}, {"loss": 1.8084, "grad_norm": 0.30274903774261475, "learning_rate": 0.0002, "epoch": 0.44237485448195574, "step": 570}, {"loss": 1.7673, "grad_norm": 0.3231128454208374, "learning_rate": 0.0002, "epoch": 0.45013581684128834, "step": 580}, {"loss": 1.7848, "grad_norm": 0.3255121409893036, "learning_rate": 0.0002, "epoch": 0.4578967792006209, "step": 590}, {"loss": 1.8227, "grad_norm": 0.30147507786750793, "learning_rate": 0.0002, "epoch": 0.46565774155995343, "step": 600}, {"loss": 1.7572, "grad_norm": 0.29781386256217957, "learning_rate": 0.0002, "epoch": 0.473418703919286, "step": 610}, {"loss": 1.8307, "grad_norm": 0.30914685130119324, "learning_rate": 0.0002, "epoch": 0.4811796662786185, "step": 620}, {"loss": 1.805, "grad_norm": 0.3110593855381012, "learning_rate": 0.0002, "epoch": 0.4889406286379511, "step": 630}, {"loss": 1.8228, "grad_norm": 0.3298132121562958, "learning_rate": 0.0002, "epoch": 0.49670159099728367, "step": 640}, {"loss": 1.7816, "grad_norm": 0.322122186422348, "learning_rate": 0.0002, "epoch": 0.5044625533566163, "step": 650}, {"loss": 1.8001, "grad_norm": 0.3504371643066406, "learning_rate": 0.0002, "epoch": 0.5122235157159488, "step": 660}, {"loss": 1.8682, "grad_norm": 0.3102182149887085, "learning_rate": 0.0002, "epoch": 0.5199844780752814, "step": 670}, {"loss": 1.7494, "grad_norm": 0.6113658547401428, "learning_rate": 0.0002, "epoch": 0.5277454404346139, "step": 680}, {"loss": 1.7096, "grad_norm": 0.31841862201690674, "learning_rate": 0.0002, "epoch": 0.5355064027939465, "step": 690}, {"loss": 1.7587, "grad_norm": 0.2830526530742645, "learning_rate": 0.0002, "epoch": 0.543267365153279, "step": 700}, {"loss": 1.7887, "grad_norm": 0.3048769533634186, "learning_rate": 0.0002, "epoch": 0.5510283275126115, "step": 710}, {"loss": 1.8416, "grad_norm": 0.2719033658504486, "learning_rate": 0.0002, "epoch": 0.5587892898719441, "step": 720}, {"loss": 1.786, "grad_norm": 0.3176722526550293, "learning_rate": 0.0002, "epoch": 0.5665502522312766, "step": 730}, {"loss": 1.7127, "grad_norm": 0.32491734623908997, "learning_rate": 0.0002, "epoch": 0.5743112145906092, "step": 740}, {"loss": 1.7892, "grad_norm": 0.32746851444244385, "learning_rate": 0.0002, "epoch": 0.5820721769499418, "step": 750}, {"loss": 1.7811, "grad_norm": 0.3055773973464966, "learning_rate": 0.0002, "epoch": 0.5898331393092744, "step": 760}, {"loss": 1.8597, "grad_norm": 0.30671584606170654, "learning_rate": 0.0002, "epoch": 0.5975941016686069, "step": 770}, {"loss": 1.7728, "grad_norm": 0.28770264983177185, "learning_rate": 0.0002, "epoch": 0.6053550640279395, "step": 780}, {"loss": 1.7025, "grad_norm": 0.2814285457134247, "learning_rate": 0.0002, "epoch": 0.613116026387272, "step": 790}, {"loss": 1.819, "grad_norm": 0.31554412841796875, "learning_rate": 0.0002, "epoch": 0.6208769887466046, "step": 800}, {"loss": 1.8335, "grad_norm": 0.2984226942062378, "learning_rate": 0.0002, "epoch": 0.6286379511059371, "step": 810}, {"loss": 1.7728, "grad_norm": 0.2859906554222107, "learning_rate": 0.0002, "epoch": 0.6363989134652697, "step": 820}, {"loss": 1.7408, "grad_norm": 0.2887928783893585, "learning_rate": 0.0002, "epoch": 0.6441598758246022, "step": 830}, {"loss": 1.8071, "grad_norm": 0.31287339329719543, "learning_rate": 0.0002, "epoch": 0.6519208381839348, "step": 840}, {"loss": 1.7673, "grad_norm": 0.32064181566238403, "learning_rate": 0.0002, "epoch": 0.6596818005432674, "step": 850}, {"loss": 1.7849, "grad_norm": 0.290981650352478, "learning_rate": 0.0002, "epoch": 0.6674427629026, "step": 860}, {"loss": 1.8089, "grad_norm": 0.33060121536254883, "learning_rate": 0.0002, "epoch": 0.6752037252619325, "step": 870}, {"loss": 1.7357, "grad_norm": 0.27032899856567383, "learning_rate": 0.0002, "epoch": 0.682964687621265, "step": 880}, {"loss": 1.8423, "grad_norm": 0.29031234979629517, "learning_rate": 0.0002, "epoch": 0.6907256499805976, "step": 890}, {"loss": 1.835, "grad_norm": 0.2845142185688019, "learning_rate": 0.0002, "epoch": 0.6984866123399301, "step": 900}, {"loss": 1.77, "grad_norm": 0.8638312816619873, "learning_rate": 0.0002, "epoch": 0.7062475746992627, "step": 910}, {"loss": 1.7757, "grad_norm": 0.3086668848991394, "learning_rate": 0.0002, "epoch": 0.7140085370585952, "step": 920}, {"loss": 1.7541, "grad_norm": 0.2724177837371826, "learning_rate": 0.0002, "epoch": 0.7217694994179278, "step": 930}, {"loss": 1.816, "grad_norm": 0.289559006690979, "learning_rate": 0.0002, "epoch": 0.7295304617772603, "step": 940}, {"loss": 1.7654, "grad_norm": 0.3000658452510834, "learning_rate": 0.0002, "epoch": 0.737291424136593, "step": 950}, {"loss": 1.7736, "grad_norm": 0.33544042706489563, "learning_rate": 0.0002, "epoch": 0.7450523864959255, "step": 960}, {"loss": 1.6979, "grad_norm": 0.28593236207962036, "learning_rate": 0.0002, "epoch": 0.7528133488552581, "step": 970}, {"loss": 1.8583, "grad_norm": 0.313634991645813, "learning_rate": 0.0002, "epoch": 0.7605743112145906, "step": 980}, {"loss": 1.7473, "grad_norm": 0.2949385941028595, "learning_rate": 0.0002, "epoch": 0.7683352735739232, "step": 990}, {"loss": 1.8689, "grad_norm": 0.2920108437538147, "learning_rate": 0.0002, "epoch": 0.7760962359332557, "step": 1000}, {"loss": 1.8401, "grad_norm": 0.3245100677013397, "learning_rate": 0.0002, "epoch": 0.7838571982925883, "step": 1010}, {"loss": 1.7109, "grad_norm": 0.3007619380950928, "learning_rate": 0.0002, "epoch": 0.7916181606519208, "step": 1020}, {"loss": 1.7427, "grad_norm": 0.3630852997303009, "learning_rate": 0.0002, "epoch": 0.7993791230112534, "step": 1030}, {"loss": 1.7655, "grad_norm": 0.2856379747390747, "learning_rate": 0.0002, "epoch": 0.8071400853705859, "step": 1040}, {"loss": 1.8371, "grad_norm": 0.32476478815078735, "learning_rate": 0.0002, "epoch": 0.8149010477299186, "step": 1050}, {"loss": 1.8039, "grad_norm": 0.5162565112113953, "learning_rate": 0.0002, "epoch": 0.8226620100892511, "step": 1060}, {"loss": 1.8862, "grad_norm": 0.316496342420578, "learning_rate": 0.0002, "epoch": 0.8304229724485837, "step": 1070}, {"loss": 1.8023, "grad_norm": 0.31977516412734985, "learning_rate": 0.0002, "epoch": 0.8381839348079162, "step": 1080}, {"loss": 1.8547, "grad_norm": 0.269509494304657, "learning_rate": 0.0002, "epoch": 0.8459448971672487, "step": 1090}, {"loss": 1.7811, "grad_norm": 0.31621453166007996, "learning_rate": 0.0002, "epoch": 0.8537058595265813, "step": 1100}, {"loss": 1.739, "grad_norm": 0.2946535050868988, "learning_rate": 0.0002, "epoch": 0.8614668218859138, "step": 1110}, {"loss": 1.7511, "grad_norm": 0.3088909983634949, "learning_rate": 0.0002, "epoch": 0.8692277842452464, "step": 1120}, {"loss": 1.8228, "grad_norm": 0.33033716678619385, "learning_rate": 0.0002, "epoch": 0.8769887466045789, "step": 1130}, {"loss": 1.7912, "grad_norm": 0.2954833507537842, "learning_rate": 0.0002, "epoch": 0.8847497089639115, "step": 1140}, {"loss": 1.8394, "grad_norm": 0.2950248122215271, "learning_rate": 0.0002, "epoch": 0.8925106713232441, "step": 1150}, {"loss": 1.7068, "grad_norm": 0.296661913394928, "learning_rate": 0.0002, "epoch": 0.9002716336825767, "step": 1160}, {"loss": 1.7967, "grad_norm": 0.35451310873031616, "learning_rate": 0.0002, "epoch": 0.9080325960419092, "step": 1170}, {"loss": 1.8202, "grad_norm": 0.32705947756767273, "learning_rate": 0.0002, "epoch": 0.9157935584012418, "step": 1180}, {"loss": 1.7396, "grad_norm": 0.3333960771560669, "learning_rate": 0.0002, "epoch": 0.9235545207605743, "step": 1190}, {"loss": 1.7801, "grad_norm": 0.3042232096195221, "learning_rate": 0.0002, "epoch": 0.9313154831199069, "step": 1200}, {"loss": 1.7586, "grad_norm": 0.281553715467453, "learning_rate": 0.0002, "epoch": 0.9390764454792394, "step": 1210}, {"loss": 1.7953, "grad_norm": 0.3096391558647156, "learning_rate": 0.0002, "epoch": 0.946837407838572, "step": 1220}, {"loss": 1.7401, "grad_norm": 0.2866271734237671, "learning_rate": 0.0002, "epoch": 0.9545983701979045, "step": 1230}, {"loss": 1.7211, "grad_norm": 0.28394097089767456, "learning_rate": 0.0002, "epoch": 0.962359332557237, "step": 1240}, {"loss": 1.7363, "grad_norm": 0.3249266743659973, "learning_rate": 0.0002, "epoch": 0.9701202949165697, "step": 1250}, {"loss": 1.7563, "grad_norm": 0.2896869480609894, "learning_rate": 0.0002, "epoch": 0.9778812572759022, "step": 1260}, {"loss": 1.6389, "grad_norm": 0.29224586486816406, "learning_rate": 0.0002, "epoch": 0.9856422196352348, "step": 1270}, {"loss": 1.7111, "grad_norm": 0.2820223569869995, "learning_rate": 0.0002, "epoch": 0.9934031819945673, "step": 1280}, {"eval_loss": 1.8081045150756836, "eval_runtime": 102.3056, "eval_samples_per_second": 4.956, "eval_steps_per_second": 0.626, "epoch": 0.9996119518820333, "step": 1288}, {"loss": 1.7518, "grad_norm": 0.3282551169395447, "learning_rate": 0.0002, "epoch": 1.0011641443538999, "step": 1290}, {"loss": 1.6806, "grad_norm": 0.30217495560646057, "learning_rate": 0.0002, "epoch": 1.0089251067132325, "step": 1300}, {"loss": 1.6777, "grad_norm": 0.30801767110824585, "learning_rate": 0.0002, "epoch": 1.016686069072565, "step": 1310}, {"loss": 1.7756, "grad_norm": 0.31816792488098145, "learning_rate": 0.0002, "epoch": 1.0244470314318976, "step": 1320}, {"loss": 1.6986, "grad_norm": 0.27794334292411804, "learning_rate": 0.0002, "epoch": 1.03220799379123, "step": 1330}, {"loss": 1.6931, "grad_norm": 0.3018926680088043, "learning_rate": 0.0002, "epoch": 1.0399689561505627, "step": 1340}, {"loss": 1.7033, "grad_norm": 0.3552975356578827, "learning_rate": 0.0002, "epoch": 1.0477299185098952, "step": 1350}, {"loss": 1.6782, "grad_norm": 0.32590144872665405, "learning_rate": 0.0002, "epoch": 1.0554908808692278, "step": 1360}, {"loss": 1.6479, "grad_norm": 0.3435460925102234, "learning_rate": 0.0002, "epoch": 1.0632518432285603, "step": 1370}, {"loss": 1.7451, "grad_norm": 0.35037797689437866, "learning_rate": 0.0002, "epoch": 1.071012805587893, "step": 1380}, {"loss": 1.7868, "grad_norm": 0.31398263573646545, "learning_rate": 0.0002, "epoch": 1.0787737679472253, "step": 1390}, {"loss": 1.6729, "grad_norm": 0.3134010434150696, "learning_rate": 0.0002, "epoch": 1.086534730306558, "step": 1400}, {"loss": 1.751, "grad_norm": 0.4599704444408417, "learning_rate": 0.0002, "epoch": 1.0942956926658907, "step": 1410}, {"loss": 1.6871, "grad_norm": 0.35852891206741333, "learning_rate": 0.0002, "epoch": 1.102056655025223, "step": 1420}, {"loss": 1.7083, "grad_norm": 0.35628634691238403, "learning_rate": 0.0002, "epoch": 1.1098176173845558, "step": 1430}, {"loss": 1.6166, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.1175785797438882, "step": 1440}, {"loss": 1.7344, "grad_norm": 1.3712416887283325, "learning_rate": 0.0002, "epoch": 1.1253395421032208, "step": 1450}, {"loss": 1.6542, "grad_norm": 0.38406670093536377, "learning_rate": 0.0002, "epoch": 1.1331005044625533, "step": 1460}, {"loss": 1.7104, "grad_norm": 0.3402116000652313, "learning_rate": 0.0002, "epoch": 1.140861466821886, "step": 1470}, {"loss": 1.7074, "grad_norm": 0.341189444065094, "learning_rate": 0.0002, "epoch": 1.1486224291812184, "step": 1480}, {"loss": 1.6468, "grad_norm": 0.36629995703697205, "learning_rate": 0.0002, "epoch": 1.156383391540551, "step": 1490}, {"loss": 1.6952, "grad_norm": 0.3499569296836853, "learning_rate": 0.0002, "epoch": 1.1641443538998835, "step": 1500}, {"loss": 1.6625, "grad_norm": 0.3663063943386078, "learning_rate": 0.0002, "epoch": 1.1719053162592161, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.34851500391960144, "learning_rate": 0.0002, "epoch": 1.1796662786185488, "step": 1520}, {"loss": 1.6092, "grad_norm": 0.35071656107902527, "learning_rate": 0.0002, "epoch": 1.1874272409778812, "step": 1530}, {"loss": 1.7206, "grad_norm": 0.42783796787261963, "learning_rate": 0.0002, "epoch": 1.1951882033372139, "step": 1540}, {"loss": 1.7499, "grad_norm": 0.31830692291259766, "learning_rate": 0.0002, "epoch": 1.2029491656965463, "step": 1550}, {"loss": 1.7372, "grad_norm": 0.3597424626350403, "learning_rate": 0.0002, "epoch": 1.210710128055879, "step": 1560}, {"loss": 1.6386, "grad_norm": 0.35233765840530396, "learning_rate": 0.0002, "epoch": 1.2184710904152114, "step": 1570}, {"loss": 1.6766, "grad_norm": 0.35942912101745605, "learning_rate": 0.0002, "epoch": 1.226232052774544, "step": 1580}, {"loss": 1.6598, "grad_norm": 0.36159393191337585, "learning_rate": 0.0002, "epoch": 1.2339930151338767, "step": 1590}, {"loss": 1.6697, "grad_norm": 0.3328469693660736, "learning_rate": 0.0002, "epoch": 1.2417539774932091, "step": 1600}, {"loss": 1.7594, "grad_norm": 0.3089476525783539, "learning_rate": 0.0002, "epoch": 1.2495149398525418, "step": 1610}, {"loss": 1.6805, "grad_norm": 0.30947765707969666, "learning_rate": 0.0002, "epoch": 1.2572759022118742, "step": 1620}, {"loss": 1.6899, "grad_norm": 0.32154011726379395, "learning_rate": 0.0002, "epoch": 1.265036864571207, "step": 1630}, {"loss": 1.6621, "grad_norm": 0.3480297923088074, "learning_rate": 0.0002, "epoch": 1.2727978269305393, "step": 1640}, {"loss": 1.7087, "grad_norm": 0.39471694827079773, "learning_rate": 0.0002, "epoch": 1.280558789289872, "step": 1650}, {"loss": 1.7608, "grad_norm": 0.35728853940963745, "learning_rate": 0.0002, "epoch": 1.2883197516492044, "step": 1660}, {"loss": 1.7008, "grad_norm": 0.35223081707954407, "learning_rate": 0.0002, "epoch": 1.296080714008537, "step": 1670}, {"loss": 1.7253, "grad_norm": 0.3588867485523224, "learning_rate": 0.0002, "epoch": 1.3038416763678695, "step": 1680}, {"loss": 1.6505, "grad_norm": 0.3528042733669281, "learning_rate": 0.0002, "epoch": 1.3116026387272022, "step": 1690}, {"loss": 1.6945, "grad_norm": 0.35975801944732666, "learning_rate": 0.0002, "epoch": 1.3193636010865348, "step": 1700}, {"loss": 1.6631, "grad_norm": 0.36691880226135254, "learning_rate": 0.0002, "epoch": 1.3271245634458673, "step": 1710}, {"loss": 1.7593, "grad_norm": 0.3787977695465088, "learning_rate": 0.0002, "epoch": 1.3348855258052, "step": 1720}, {"loss": 1.7697, "grad_norm": 0.36614933609962463, "learning_rate": 0.0002, "epoch": 1.3426464881645324, "step": 1730}, {"loss": 1.6487, "grad_norm": 0.3484745919704437, "learning_rate": 0.0002, "epoch": 1.350407450523865, "step": 1740}, {"loss": 1.7054, "grad_norm": 0.36905673146247864, "learning_rate": 0.0002, "epoch": 1.3581684128831975, "step": 1750}, {"loss": 1.7679, "grad_norm": 0.41564738750457764, "learning_rate": 0.0002, "epoch": 1.36592937524253, "step": 1760}, {"loss": 1.6634, "grad_norm": 0.3345205783843994, "learning_rate": 0.0002, "epoch": 1.3736903376018628, "step": 1770}, {"loss": 1.7275, "grad_norm": 0.34926071763038635, "learning_rate": 0.0002, "epoch": 1.3814512999611952, "step": 1780}, {"loss": 1.685, "grad_norm": 0.42004233598709106, "learning_rate": 0.0002, "epoch": 1.3892122623205276, "step": 1790}, {"loss": 1.666, "grad_norm": 0.3576236963272095, "learning_rate": 0.0002, "epoch": 1.3969732246798603, "step": 1800}, {"loss": 1.8516, "grad_norm": 0.3586704432964325, "learning_rate": 0.0002, "epoch": 1.404734187039193, "step": 1810}, {"loss": 1.6171, "grad_norm": 0.3943439722061157, "learning_rate": 0.0002, "epoch": 1.4124951493985254, "step": 1820}, {"loss": 1.6865, "grad_norm": 0.3484877049922943, "learning_rate": 0.0002, "epoch": 1.420256111757858, "step": 1830}, {"loss": 1.7205, "grad_norm": 0.3344518840312958, "learning_rate": 0.0002, "epoch": 1.4280170741171905, "step": 1840}, {"loss": 1.6999, "grad_norm": 0.4345698356628418, "learning_rate": 0.0002, "epoch": 1.4357780364765231, "step": 1850}, {"loss": 1.6855, "grad_norm": 0.5525162220001221, "learning_rate": 0.0002, "epoch": 1.4435389988358556, "step": 1860}, {"loss": 1.7143, "grad_norm": 0.37194496393203735, "learning_rate": 0.0002, "epoch": 1.4512999611951882, "step": 1870}, {"loss": 1.7623, "grad_norm": 0.34570157527923584, "learning_rate": 0.0002, "epoch": 1.4590609235545209, "step": 1880}, {"loss": 1.7, "grad_norm": 0.3512282073497772, "learning_rate": 0.0002, "epoch": 1.4668218859138533, "step": 1890}, {"loss": 1.7225, "grad_norm": 0.3443922996520996, "learning_rate": 0.0002, "epoch": 1.4745828482731858, "step": 1900}, {"loss": 1.7393, "grad_norm": 0.3812018036842346, "learning_rate": 0.0002, "epoch": 1.4823438106325184, "step": 1910}, {"loss": 1.7277, "grad_norm": 0.39263492822647095, "learning_rate": 0.0002, "epoch": 1.490104772991851, "step": 1920}, {"loss": 1.6829, "grad_norm": 0.3146156072616577, "learning_rate": 0.0002, "epoch": 1.4978657353511835, "step": 1930}, {"loss": 1.6881, "grad_norm": 0.3653988540172577, "learning_rate": 0.0002, "epoch": 1.505626697710516, "step": 1940}, {"loss": 1.7064, "grad_norm": 0.3966596722602844, "learning_rate": 0.0002, "epoch": 1.5133876600698488, "step": 1950}, {"loss": 1.6942, "grad_norm": 0.3441697359085083, "learning_rate": 0.0002, "epoch": 1.5211486224291813, "step": 1960}, {"loss": 1.7175, "grad_norm": 0.3328564465045929, "learning_rate": 0.0002, "epoch": 1.5289095847885137, "step": 1970}, {"loss": 1.7394, "grad_norm": 0.34068772196769714, "learning_rate": 0.0002, "epoch": 1.5366705471478463, "step": 1980}, {"loss": 1.7016, "grad_norm": 0.3559795916080475, "learning_rate": 0.0002, "epoch": 1.544431509507179, "step": 1990}, {"loss": 1.7102, "grad_norm": 0.37888768315315247, "learning_rate": 0.0002, "epoch": 1.5521924718665114, "step": 2000}, {"loss": 1.7094, "grad_norm": 0.36128363013267517, "learning_rate": 0.0002, "epoch": 1.5599534342258439, "step": 2010}, {"loss": 1.6407, "grad_norm": 0.3643714487552643, "learning_rate": 0.0002, "epoch": 1.5677143965851765, "step": 2020}, {"loss": 1.6777, "grad_norm": 0.3863612115383148, "learning_rate": 0.0002, "epoch": 1.5754753589445092, "step": 2030}, {"loss": 1.6575, "grad_norm": 0.32831457257270813, "learning_rate": 0.0002, "epoch": 1.5832363213038416, "step": 2040}, {"loss": 1.7404, "grad_norm": 0.36098113656044006, "learning_rate": 0.0002, "epoch": 1.5909972836631743, "step": 2050}, {"loss": 1.7065, "grad_norm": 1.1079334020614624, "learning_rate": 0.0002, "epoch": 1.598758246022507, "step": 2060}, {"loss": 1.6824, "grad_norm": 0.35615381598472595, "learning_rate": 0.0002, "epoch": 1.6065192083818394, "step": 2070}, {"loss": 1.7262, "grad_norm": 0.369711309671402, "learning_rate": 0.0002, "epoch": 1.6142801707411718, "step": 2080}, {"loss": 1.6995, "grad_norm": 0.390658438205719, "learning_rate": 0.0002, "epoch": 1.6220411331005045, "step": 2090}, {"loss": 1.6996, "grad_norm": 0.3422999382019043, "learning_rate": 0.0002, "epoch": 1.6298020954598371, "step": 2100}, {"loss": 1.7135, "grad_norm": 0.372475266456604, "learning_rate": 0.0002, "epoch": 1.6375630578191696, "step": 2110}, {"loss": 1.7216, "grad_norm": 0.35660576820373535, "learning_rate": 0.0002, "epoch": 1.645324020178502, "step": 2120}, {"loss": 1.6991, "grad_norm": 0.35754942893981934, "learning_rate": 0.0002, "epoch": 1.6530849825378346, "step": 2130}, {"loss": 1.6779, "grad_norm": 0.34572410583496094, "learning_rate": 0.0002, "epoch": 1.6608459448971673, "step": 2140}, {"loss": 1.6707, "grad_norm": 0.42059701681137085, "learning_rate": 0.0002, "epoch": 1.6686069072564997, "step": 2150}, {"loss": 1.6782, "grad_norm": 0.35200759768486023, "learning_rate": 0.0002, "epoch": 1.6763678696158324, "step": 2160}, {"loss": 1.6869, "grad_norm": 0.3704029321670532, "learning_rate": 0.0002, "epoch": 1.684128831975165, "step": 2170}, {"loss": 1.7192, "grad_norm": 0.40450501441955566, "learning_rate": 0.0002, "epoch": 1.6918897943344975, "step": 2180}, {"loss": 1.6228, "grad_norm": 0.362966924905777, "learning_rate": 0.0002, "epoch": 1.69965075669383, "step": 2190}, {"loss": 1.6935, "grad_norm": 0.36586204171180725, "learning_rate": 0.0002, "epoch": 1.7074117190531626, "step": 2200}, {"loss": 1.6088, "grad_norm": 0.3295372426509857, "learning_rate": 0.0002, "epoch": 1.7151726814124952, "step": 2210}, {"loss": 1.7844, "grad_norm": 0.3892575800418854, "learning_rate": 0.0002, "epoch": 1.7229336437718277, "step": 2220}, {"loss": 1.7805, "grad_norm": 0.34712135791778564, "learning_rate": 0.0002, "epoch": 1.73069460613116, "step": 2230}, {"loss": 1.7353, "grad_norm": 0.34801796078681946, "learning_rate": 0.0002, "epoch": 1.738455568490493, "step": 2240}, {"loss": 1.7009, "grad_norm": 0.3822397291660309, "learning_rate": 0.0002, "epoch": 1.7462165308498254, "step": 2250}, {"loss": 1.6546, "grad_norm": 0.38933250308036804, "learning_rate": 0.0002, "epoch": 1.7539774932091579, "step": 2260}, {"loss": 1.7245, "grad_norm": 0.3798373341560364, "learning_rate": 0.0002, "epoch": 1.7617384555684905, "step": 2270}, {"loss": 1.6508, "grad_norm": 0.35151317715644836, "learning_rate": 0.0002, "epoch": 1.7694994179278232, "step": 2280}, {"loss": 1.6894, "grad_norm": 0.44981494545936584, "learning_rate": 0.0002, "epoch": 1.7772603802871556, "step": 2290}, {"loss": 1.7271, "grad_norm": 0.3992624580860138, "learning_rate": 0.0002, "epoch": 1.785021342646488, "step": 2300}, {"loss": 1.7252, "grad_norm": 0.3772512376308441, "learning_rate": 0.0002, "epoch": 1.7927823050058207, "step": 2310}, {"loss": 1.7057, "grad_norm": 0.3511589467525482, "learning_rate": 0.0002, "epoch": 1.8005432673651534, "step": 2320}, {"loss": 1.764, "grad_norm": 0.3805285394191742, "learning_rate": 0.0002, "epoch": 1.8083042297244858, "step": 2330}, {"loss": 1.6986, "grad_norm": 0.3792071044445038, "learning_rate": 0.0002, "epoch": 1.8160651920838184, "step": 2340}, {"loss": 1.7759, "grad_norm": 0.36430829763412476, "learning_rate": 0.0002, "epoch": 1.823826154443151, "step": 2350}, {"loss": 1.6773, "grad_norm": 0.36502477526664734, "learning_rate": 0.0002, "epoch": 1.8315871168024835, "step": 2360}, {"loss": 1.8072, "grad_norm": 0.35015153884887695, "learning_rate": 0.0002, "epoch": 1.839348079161816, "step": 2370}, {"loss": 1.7734, "grad_norm": 0.3710903823375702, "learning_rate": 0.0002, "epoch": 1.8471090415211486, "step": 2380}, {"loss": 1.6737, "grad_norm": 0.3542828857898712, "learning_rate": 0.0002, "epoch": 1.8548700038804813, "step": 2390}, {"loss": 1.6783, "grad_norm": 0.35467568039894104, "learning_rate": 0.0002, "epoch": 1.8626309662398137, "step": 2400}, {"loss": 1.7773, "grad_norm": 0.3638560473918915, "learning_rate": 0.0002, "epoch": 1.8703919285991462, "step": 2410}, {"loss": 1.7019, "grad_norm": 0.3823298215866089, "learning_rate": 0.0002, "epoch": 1.8781528909584788, "step": 2420}, {"loss": 1.6935, "grad_norm": 0.3926416337490082, "learning_rate": 0.0002, "epoch": 1.8859138533178115, "step": 2430}, {"loss": 1.71, "grad_norm": 0.3608079254627228, "learning_rate": 0.0002, "epoch": 1.893674815677144, "step": 2440}, {"loss": 1.6654, "grad_norm": 0.3426613509654999, "learning_rate": 0.0002, "epoch": 1.9014357780364766, "step": 2450}, {"loss": 1.6892, "grad_norm": 0.3522338569164276, "learning_rate": 0.0002, "epoch": 1.9091967403958092, "step": 2460}, {"loss": 1.7307, "grad_norm": 0.3608049154281616, "learning_rate": 0.0002, "epoch": 1.9169577027551417, "step": 2470}, {"loss": 1.6823, "grad_norm": 0.3849755525588989, "learning_rate": 0.0002, "epoch": 1.924718665114474, "step": 2480}, {"loss": 1.7518, "grad_norm": 0.4154011011123657, "learning_rate": 0.0002, "epoch": 1.9324796274738067, "step": 2490}, {"loss": 1.7381, "grad_norm": 0.3602796792984009, "learning_rate": 0.0002, "epoch": 1.9402405898331394, "step": 2500}, {"loss": 1.7843, "grad_norm": 0.3702992796897888, "learning_rate": 0.0002, "epoch": 1.9480015521924718, "step": 2510}, {"loss": 1.6669, "grad_norm": 0.3657735288143158, "learning_rate": 0.0002, "epoch": 1.9557625145518043, "step": 2520}, {"loss": 1.5964, "grad_norm": 0.41031739115715027, "learning_rate": 0.0002, "epoch": 1.963523476911137, "step": 2530}, {"loss": 1.6745, "grad_norm": 0.34578680992126465, "learning_rate": 0.0002, "epoch": 1.9712844392704696, "step": 2540}, {"loss": 1.723, "grad_norm": 0.3361521065235138, "learning_rate": 0.0002, "epoch": 1.979045401629802, "step": 2550}, {"loss": 1.6868, "grad_norm": 0.34342363476753235, "learning_rate": 0.0002, "epoch": 1.9868063639891347, "step": 2560}, {"loss": 1.6577, "grad_norm": 0.32954007387161255, "learning_rate": 0.0002, "epoch": 1.9945673263484673, "step": 2570}, {"eval_loss": 1.8068748712539673, "eval_runtime": 105.5885, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 2.0, "step": 2577}, {"loss": 1.634, "grad_norm": 0.336302250623703, "learning_rate": 0.0002, "epoch": 2.0023282887077998, "step": 2580}, {"loss": 1.612, "grad_norm": 0.3627048432826996, "learning_rate": 0.0002, "epoch": 2.010089251067132, "step": 2590}, {"loss": 1.4908, "grad_norm": 0.38406702876091003, "learning_rate": 0.0002, "epoch": 2.017850213426465, "step": 2600}, {"loss": 1.5368, "grad_norm": 0.5326781272888184, "learning_rate": 0.0002, "epoch": 2.0256111757857975, "step": 2610}, {"loss": 1.5727, "grad_norm": 0.4774554967880249, "learning_rate": 0.0002, "epoch": 2.03337213814513, "step": 2620}, {"loss": 1.5422, "grad_norm": 0.4251810312271118, "learning_rate": 0.0002, "epoch": 2.0411331005044624, "step": 2630}, {"loss": 1.5152, "grad_norm": 0.4693007171154022, "learning_rate": 0.0002, "epoch": 2.0488940628637953, "step": 2640}, {"loss": 1.6137, "grad_norm": 0.46371519565582275, "learning_rate": 0.0002, "epoch": 2.0566550252231277, "step": 2650}, {"loss": 1.6304, "grad_norm": 0.46652570366859436, "learning_rate": 0.0002, "epoch": 2.06441598758246, "step": 2660}, {"loss": 1.6022, "grad_norm": 0.45200315117836, "learning_rate": 0.0002, "epoch": 2.0721769499417926, "step": 2670}, {"loss": 1.5358, "grad_norm": 0.42905205488204956, "learning_rate": 0.0002, "epoch": 2.0799379123011255, "step": 2680}, {"loss": 1.5401, "grad_norm": 0.44509148597717285, "learning_rate": 0.0002, "epoch": 2.087698874660458, "step": 2690}, {"loss": 1.5303, "grad_norm": 0.4445319175720215, "learning_rate": 0.0002, "epoch": 2.0954598370197903, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.46825504302978516, "learning_rate": 0.0002, "epoch": 2.103220799379123, "step": 2710}, {"loss": 1.5751, "grad_norm": 0.4623856842517853, "learning_rate": 0.0002, "epoch": 2.1109817617384556, "step": 2720}, {"loss": 1.5601, "grad_norm": 0.4833452105522156, "learning_rate": 0.0002, "epoch": 2.118742724097788, "step": 2730}, {"loss": 1.5997, "grad_norm": 0.4582686722278595, "learning_rate": 0.0002, "epoch": 2.1265036864571205, "step": 2740}, {"loss": 1.5801, "grad_norm": 0.47587934136390686, "learning_rate": 0.0002, "epoch": 2.1342646488164534, "step": 2750}, {"loss": 1.594, "grad_norm": 0.4602217972278595, "learning_rate": 0.0002, "epoch": 2.142025611175786, "step": 2760}, {"loss": 1.5271, "grad_norm": 0.47501352429389954, "learning_rate": 0.0002, "epoch": 2.1497865735351183, "step": 2770}, {"loss": 1.4862, "grad_norm": 0.5078499913215637, "learning_rate": 0.0002, "epoch": 2.1575475358944507, "step": 2780}, {"loss": 1.6236, "grad_norm": 0.497704416513443, "learning_rate": 0.0002, "epoch": 2.1653084982537836, "step": 2790}, {"loss": 1.5597, "grad_norm": 0.5435971617698669, "learning_rate": 0.0002, "epoch": 2.173069460613116, "step": 2800}, {"loss": 1.5926, "grad_norm": 0.5172356367111206, "learning_rate": 0.0002, "epoch": 2.1808304229724484, "step": 2810}, {"loss": 1.5202, "grad_norm": 0.44063422083854675, "learning_rate": 0.0002, "epoch": 2.1885913853317813, "step": 2820}, {"loss": 1.6041, "grad_norm": 0.5079569220542908, "learning_rate": 0.0002, "epoch": 2.1963523476911138, "step": 2830}, {"loss": 1.5915, "grad_norm": 0.45658132433891296, "learning_rate": 0.0002, "epoch": 2.204113310050446, "step": 2840}, {"loss": 1.5546, "grad_norm": 0.5103023648262024, "learning_rate": 0.0002, "epoch": 2.2118742724097786, "step": 2850}, {"loss": 1.6197, "grad_norm": 0.4882226288318634, "learning_rate": 0.0002, "epoch": 2.2196352347691115, "step": 2860}, {"loss": 1.5996, "grad_norm": 0.5087296962738037, "learning_rate": 0.0002, "epoch": 2.227396197128444, "step": 2870}, {"loss": 1.5451, "grad_norm": 0.45293712615966797, "learning_rate": 0.0002, "epoch": 2.2351571594877764, "step": 2880}, {"loss": 1.6214, "grad_norm": 0.5120379328727722, "learning_rate": 0.0002, "epoch": 2.242918121847109, "step": 2890}, {"loss": 1.5273, "grad_norm": 0.47126415371894836, "learning_rate": 0.0002, "epoch": 2.2506790842064417, "step": 2900}, {"loss": 1.612, "grad_norm": 0.44005846977233887, "learning_rate": 0.0002, "epoch": 2.258440046565774, "step": 2910}, {"loss": 1.6023, "grad_norm": 0.46476176381111145, "learning_rate": 0.0002, "epoch": 2.2662010089251066, "step": 2920}, {"loss": 1.6417, "grad_norm": 0.48051515221595764, "learning_rate": 0.0002, "epoch": 2.2739619712844394, "step": 2930}, {"loss": 1.587, "grad_norm": 0.480069637298584, "learning_rate": 0.0002, "epoch": 2.281722933643772, "step": 2940}, {"loss": 1.5747, "grad_norm": 0.5122102499008179, "learning_rate": 0.0002, "epoch": 2.2894838960031043, "step": 2950}, {"loss": 1.5183, "grad_norm": 0.48879891633987427, "learning_rate": 0.0002, "epoch": 2.2972448583624367, "step": 2960}, {"loss": 1.5483, "grad_norm": 0.4973136782646179, "learning_rate": 0.0002, "epoch": 2.3050058207217696, "step": 2970}, {"loss": 1.677, "grad_norm": 0.5522695183753967, "learning_rate": 0.0002, "epoch": 2.312766783081102, "step": 2980}, {"loss": 1.5946, "grad_norm": 0.5220217704772949, "learning_rate": 0.0002, "epoch": 2.3205277454404345, "step": 2990}, {"loss": 1.6299, "grad_norm": 0.4978662431240082, "learning_rate": 0.0002, "epoch": 2.328288707799767, "step": 3000}, {"loss": 1.5498, "grad_norm": 0.554053544998169, "learning_rate": 0.0002, "epoch": 2.3360496701591, "step": 3010}, {"loss": 1.5356, "grad_norm": 0.4703886806964874, "learning_rate": 0.0002, "epoch": 2.3438106325184322, "step": 3020}, {"loss": 1.5418, "grad_norm": 0.5074123740196228, "learning_rate": 0.0002, "epoch": 2.3515715948777647, "step": 3030}, {"loss": 1.6873, "grad_norm": 0.5088278651237488, "learning_rate": 0.0002, "epoch": 2.3593325572370976, "step": 3040}, {"loss": 1.5249, "grad_norm": 0.4752114415168762, "learning_rate": 0.0002, "epoch": 2.36709351959643, "step": 3050}, {"loss": 1.5353, "grad_norm": 0.5121659636497498, "learning_rate": 0.0002, "epoch": 2.3748544819557624, "step": 3060}, {"loss": 1.6426, "grad_norm": 0.48649218678474426, "learning_rate": 0.0002, "epoch": 2.3826154443150953, "step": 3070}, {"loss": 1.6136, "grad_norm": 0.5209488868713379, "learning_rate": 0.0002, "epoch": 2.3903764066744277, "step": 3080}, {"loss": 1.597, "grad_norm": 0.5110517740249634, "learning_rate": 0.0002, "epoch": 2.39813736903376, "step": 3090}, {"loss": 1.5773, "grad_norm": 0.5609337091445923, "learning_rate": 0.0002, "epoch": 2.4058983313930926, "step": 3100}, {"loss": 1.5438, "grad_norm": 0.5191826224327087, "learning_rate": 0.0002, "epoch": 2.4136592937524255, "step": 3110}, {"loss": 1.6347, "grad_norm": 0.4876069724559784, "learning_rate": 0.0002, "epoch": 2.421420256111758, "step": 3120}, {"loss": 1.5565, "grad_norm": 0.4713933765888214, "learning_rate": 0.0002, "epoch": 2.4291812184710904, "step": 3130}, {"loss": 1.6388, "grad_norm": 0.5102227330207825, "learning_rate": 0.0002, "epoch": 2.436942180830423, "step": 3140}, {"loss": 1.5667, "grad_norm": 0.44546666741371155, "learning_rate": 0.0002, "epoch": 2.4447031431897557, "step": 3150}, {"loss": 1.5973, "grad_norm": 0.5167558193206787, "learning_rate": 0.0002, "epoch": 2.452464105549088, "step": 3160}, {"loss": 1.5673, "grad_norm": 0.5226958990097046, "learning_rate": 0.0002, "epoch": 2.4602250679084205, "step": 3170}, {"loss": 1.5758, "grad_norm": 0.4751799702644348, "learning_rate": 0.0002, "epoch": 2.4679860302677534, "step": 3180}, {"loss": 1.6234, "grad_norm": 0.4744729697704315, "learning_rate": 0.0002, "epoch": 2.475746992627086, "step": 3190}, {"loss": 1.5661, "grad_norm": 0.5203230381011963, "learning_rate": 0.0002, "epoch": 2.4835079549864183, "step": 3200}, {"loss": 1.493, "grad_norm": 0.47209781408309937, "learning_rate": 0.0002, "epoch": 2.4912689173457507, "step": 3210}, {"loss": 1.6415, "grad_norm": 0.5241674780845642, "learning_rate": 0.0002, "epoch": 2.4990298797050836, "step": 3220}, {"loss": 1.6324, "grad_norm": 0.5152244567871094, "learning_rate": 0.0002, "epoch": 2.506790842064416, "step": 3230}, {"loss": 1.6248, "grad_norm": 0.5216741561889648, "learning_rate": 0.0002, "epoch": 2.5145518044237485, "step": 3240}, {"loss": 1.5668, "grad_norm": 0.4953259527683258, "learning_rate": 0.0002, "epoch": 2.522312766783081, "step": 3250}, {"loss": 1.666, "grad_norm": 0.5973829030990601, "learning_rate": 0.0002, "epoch": 2.530073729142414, "step": 3260}, {"loss": 1.5295, "grad_norm": 0.48804202675819397, "learning_rate": 0.0002, "epoch": 2.5378346915017462, "step": 3270}, {"loss": 1.4954, "grad_norm": 0.5334644317626953, "learning_rate": 0.0002, "epoch": 2.5455956538610787, "step": 3280}, {"loss": 1.5814, "grad_norm": 0.46873313188552856, "learning_rate": 0.0002, "epoch": 2.5533566162204115, "step": 3290}, {"loss": 1.5362, "grad_norm": 0.4282589554786682, "learning_rate": 0.0002, "epoch": 2.561117578579744, "step": 3300}, {"loss": 1.6278, "grad_norm": 0.4848293960094452, "learning_rate": 0.0002, "epoch": 2.5688785409390764, "step": 3310}, {"loss": 1.6308, "grad_norm": 0.5093745589256287, "learning_rate": 0.0002, "epoch": 2.576639503298409, "step": 3320}, {"loss": 1.6375, "grad_norm": 0.5084842443466187, "learning_rate": 0.0002, "epoch": 2.5844004656577413, "step": 3330}, {"loss": 1.6168, "grad_norm": 0.4696281850337982, "learning_rate": 0.0002, "epoch": 2.592161428017074, "step": 3340}, {"loss": 1.5359, "grad_norm": 0.5767765641212463, "learning_rate": 0.0002, "epoch": 2.5999223903764066, "step": 3350}, {"loss": 1.6097, "grad_norm": 0.47300875186920166, "learning_rate": 0.0002, "epoch": 2.607683352735739, "step": 3360}, {"loss": 1.6138, "grad_norm": 0.4809158146381378, "learning_rate": 0.0002, "epoch": 2.615444315095072, "step": 3370}, {"loss": 1.4952, "grad_norm": 0.5141063928604126, "learning_rate": 0.0002, "epoch": 2.6232052774544043, "step": 3380}, {"loss": 1.5784, "grad_norm": 0.4832935035228729, "learning_rate": 0.0002, "epoch": 2.630966239813737, "step": 3390}, {"loss": 1.5796, "grad_norm": 0.5044625401496887, "learning_rate": 0.0002, "epoch": 2.6387272021730697, "step": 3400}, {"loss": 1.6202, "grad_norm": 0.5287680625915527, "learning_rate": 0.0002, "epoch": 2.646488164532402, "step": 3410}, {"loss": 1.5423, "grad_norm": 0.5306379795074463, "learning_rate": 0.0002, "epoch": 2.6542491268917345, "step": 3420}, {"loss": 1.5264, "grad_norm": 0.5849291682243347, "learning_rate": 0.0002, "epoch": 2.662010089251067, "step": 3430}, {"loss": 1.5937, "grad_norm": 0.7951080799102783, "learning_rate": 0.0002, "epoch": 2.6697710516104, "step": 3440}, {"loss": 1.5791, "grad_norm": 0.48087653517723083, "learning_rate": 0.0002, "epoch": 2.6775320139697323, "step": 3450}, {"loss": 1.6769, "grad_norm": 0.5396431684494019, "learning_rate": 0.0002, "epoch": 2.6852929763290647, "step": 3460}, {"loss": 1.606, "grad_norm": 0.5481634736061096, "learning_rate": 0.0002, "epoch": 2.693053938688397, "step": 3470}, {"loss": 1.6436, "grad_norm": 0.5068731307983398, "learning_rate": 0.0002, "epoch": 2.70081490104773, "step": 3480}, {"loss": 1.5738, "grad_norm": 0.5759826898574829, "learning_rate": 0.0002, "epoch": 2.7085758634070625, "step": 3490}, {"loss": 1.596, "grad_norm": 0.7253932952880859, "learning_rate": 0.0002, "epoch": 2.716336825766395, "step": 3500}, {"loss": 1.5791, "grad_norm": 0.527745246887207, "learning_rate": 0.0002, "epoch": 2.724097788125728, "step": 3510}, {"loss": 1.5874, "grad_norm": 0.5279242396354675, "learning_rate": 0.0002, "epoch": 2.73185875048506, "step": 3520}, {"loss": 1.6768, "grad_norm": 0.5047839283943176, "learning_rate": 0.0002, "epoch": 2.7396197128443927, "step": 3530}, {"loss": 1.5517, "grad_norm": 0.5430883169174194, "learning_rate": 0.0002, "epoch": 2.7473806752037255, "step": 3540}, {"loss": 1.5624, "grad_norm": 0.4496723711490631, "learning_rate": 0.0002, "epoch": 2.755141637563058, "step": 3550}, {"loss": 1.5789, "grad_norm": 0.5063338875770569, "learning_rate": 0.0002, "epoch": 2.7629025999223904, "step": 3560}, {"loss": 1.52, "grad_norm": 0.4619026780128479, "learning_rate": 0.0002, "epoch": 2.770663562281723, "step": 3570}, {"loss": 1.5793, "grad_norm": 0.4753304123878479, "learning_rate": 0.0002, "epoch": 2.7784245246410553, "step": 3580}, {"loss": 1.5715, "grad_norm": 0.5422708988189697, "learning_rate": 0.0002, "epoch": 2.786185487000388, "step": 3590}, {"loss": 1.5926, "grad_norm": 0.4756578803062439, "learning_rate": 0.0002, "epoch": 2.7939464493597206, "step": 3600}, {"loss": 1.5358, "grad_norm": 0.5057567358016968, "learning_rate": 0.0002, "epoch": 2.801707411719053, "step": 3610}, {"loss": 1.6131, "grad_norm": 0.5410919785499573, "learning_rate": 0.0002, "epoch": 2.809468374078386, "step": 3620}, {"loss": 1.5573, "grad_norm": 0.4958136975765228, "learning_rate": 0.0002, "epoch": 2.8172293364377183, "step": 3630}, {"loss": 1.6324, "grad_norm": 0.454527348279953, "learning_rate": 0.0002, "epoch": 2.8249902987970508, "step": 3640}, {"loss": 1.5582, "grad_norm": 0.5092706084251404, "learning_rate": 0.0002, "epoch": 2.8327512611563836, "step": 3650}, {"loss": 1.5893, "grad_norm": 0.5314022302627563, "learning_rate": 0.0002, "epoch": 2.840512223515716, "step": 3660}, {"loss": 1.588, "grad_norm": 0.5028239488601685, "learning_rate": 0.0002, "epoch": 2.8482731858750485, "step": 3670}, {"loss": 1.5751, "grad_norm": 0.5127444863319397, "learning_rate": 0.0002, "epoch": 2.856034148234381, "step": 3680}, {"loss": 1.6018, "grad_norm": 0.5045645236968994, "learning_rate": 0.0002, "epoch": 2.8637951105937134, "step": 3690}, {"loss": 1.5788, "grad_norm": 0.5560781955718994, "learning_rate": 0.0002, "epoch": 2.8715560729530463, "step": 3700}, {"loss": 1.5988, "grad_norm": 0.5177600383758545, "learning_rate": 0.0002, "epoch": 2.8793170353123787, "step": 3710}, {"loss": 1.6009, "grad_norm": 0.45830899477005005, "learning_rate": 0.0002, "epoch": 2.887077997671711, "step": 3720}, {"loss": 1.6344, "grad_norm": 0.4828629195690155, "learning_rate": 0.0002, "epoch": 2.894838960031044, "step": 3730}, {"loss": 1.6758, "grad_norm": 0.48241183161735535, "learning_rate": 0.0002, "epoch": 2.9025999223903765, "step": 3740}, {"loss": 1.5649, "grad_norm": 0.4909592568874359, "learning_rate": 0.0002, "epoch": 2.910360884749709, "step": 3750}, {"loss": 1.4927, "grad_norm": 0.44677025079727173, "learning_rate": 0.0002, "epoch": 2.9181218471090418, "step": 3760}, {"loss": 1.5067, "grad_norm": 0.4928834140300751, "learning_rate": 0.0002, "epoch": 2.925882809468374, "step": 3770}, {"loss": 1.5843, "grad_norm": 0.5673553347587585, "learning_rate": 0.0002, "epoch": 2.9336437718277066, "step": 3780}, {"loss": 1.5566, "grad_norm": 0.548190712928772, "learning_rate": 0.0002, "epoch": 2.941404734187039, "step": 3790}, {"loss": 1.5892, "grad_norm": 0.48979803919792175, "learning_rate": 0.0002, "epoch": 2.9491656965463715, "step": 3800}, {"loss": 1.5589, "grad_norm": 0.533191978931427, "learning_rate": 0.0002, "epoch": 2.9569266589057044, "step": 3810}, {"loss": 1.584, "grad_norm": 0.5362946391105652, "learning_rate": 0.0002, "epoch": 2.964687621265037, "step": 3820}, {"loss": 1.6602, "grad_norm": 0.4724906384944916, "learning_rate": 0.0002, "epoch": 2.9724485836243693, "step": 3830}, {"loss": 1.5834, "grad_norm": 0.5468461513519287, "learning_rate": 0.0002, "epoch": 2.980209545983702, "step": 3840}, {"loss": 1.6316, "grad_norm": 0.4697108864784241, "learning_rate": 0.0002, "epoch": 2.9879705083430346, "step": 3850}, {"loss": 1.6312, "grad_norm": 0.4780906140804291, "learning_rate": 0.0002, "epoch": 2.995731470702367, "step": 3860}, {"eval_loss": 1.8472607135772705, "eval_runtime": 106.5541, "eval_samples_per_second": 4.758, "eval_steps_per_second": 0.601, "epoch": 2.9996119518820334, "step": 3865}, {"loss": 1.4983, "grad_norm": 0.5645653605461121, "learning_rate": 0.0002, "epoch": 3.0034924330616994, "step": 3870}, {"loss": 1.4334, "grad_norm": 0.6457151174545288, "learning_rate": 0.0002, "epoch": 3.0112533954210323, "step": 3880}, {"loss": 1.3899, "grad_norm": 0.583838164806366, "learning_rate": 0.0002, "epoch": 3.0190143577803648, "step": 3890}, {"loss": 1.3258, "grad_norm": 0.6819260120391846, "learning_rate": 0.0002, "epoch": 3.026775320139697, "step": 3900}, {"loss": 1.3458, "grad_norm": 0.6692903637886047, "learning_rate": 0.0002, "epoch": 3.03453628249903, "step": 3910}, {"loss": 1.4356, "grad_norm": 0.6101024746894836, "learning_rate": 0.0002, "epoch": 3.0422972448583625, "step": 3920}, {"loss": 1.394, "grad_norm": 0.7014093399047852, "learning_rate": 0.0002, "epoch": 3.050058207217695, "step": 3930}, {"loss": 1.3885, "grad_norm": 0.7380381226539612, "learning_rate": 0.0002, "epoch": 3.0578191695770274, "step": 3940}, {"loss": 1.4206, "grad_norm": 0.6607900857925415, "learning_rate": 0.0002, "epoch": 3.0655801319363603, "step": 3950}, {"loss": 1.4293, "grad_norm": 0.735263466835022, "learning_rate": 0.0002, "epoch": 3.0733410942956927, "step": 3960}, {"loss": 1.3966, "grad_norm": 0.6788513660430908, "learning_rate": 0.0002, "epoch": 3.081102056655025, "step": 3970}, {"loss": 1.3435, "grad_norm": 0.6347652673721313, "learning_rate": 0.0002, "epoch": 3.088863019014358, "step": 3980}, {"loss": 1.4518, "grad_norm": 0.7056642770767212, "learning_rate": 0.0002, "epoch": 3.0966239813736904, "step": 3990}, {"loss": 1.4474, "grad_norm": 0.6387075185775757, "learning_rate": 0.0002, "epoch": 3.104384943733023, "step": 4000}, {"loss": 1.3833, "grad_norm": 0.6701116561889648, "learning_rate": 0.0002, "epoch": 3.1121459060923553, "step": 4010}, {"loss": 1.404, "grad_norm": 0.7558449506759644, "learning_rate": 0.0002, "epoch": 3.119906868451688, "step": 4020}, {"loss": 1.3294, "grad_norm": 0.6612881422042847, "learning_rate": 0.0002, "epoch": 3.1276678308110206, "step": 4030}, {"loss": 1.439, "grad_norm": 0.7474587559700012, "learning_rate": 0.0002, "epoch": 3.135428793170353, "step": 4040}, {"loss": 1.4616, "grad_norm": 0.7292373776435852, "learning_rate": 0.0002, "epoch": 3.1431897555296855, "step": 4050}, {"loss": 1.3908, "grad_norm": 0.7432886958122253, "learning_rate": 0.0002, "epoch": 3.1509507178890184, "step": 4060}, {"loss": 1.4214, "grad_norm": 0.6366098523139954, "learning_rate": 0.0002, "epoch": 3.158711680248351, "step": 4070}, {"loss": 1.5044, "grad_norm": 0.6837611794471741, "learning_rate": 0.0002, "epoch": 3.1664726426076832, "step": 4080}, {"loss": 1.4332, "grad_norm": 0.7194393277168274, "learning_rate": 0.0002, "epoch": 3.174233604967016, "step": 4090}, {"loss": 1.3628, "grad_norm": 0.6963607668876648, "learning_rate": 0.0002, "epoch": 3.1819945673263486, "step": 4100}, {"loss": 1.4127, "grad_norm": 0.6404902935028076, "learning_rate": 0.0002, "epoch": 3.189755529685681, "step": 4110}, {"loss": 1.4394, "grad_norm": 0.7172070741653442, "learning_rate": 0.0002, "epoch": 3.1975164920450134, "step": 4120}, {"loss": 1.4658, "grad_norm": 0.6577759385108948, "learning_rate": 0.0002, "epoch": 3.2052774544043463, "step": 4130}, {"loss": 1.4019, "grad_norm": 0.6658480167388916, "learning_rate": 0.0002, "epoch": 3.2130384167636787, "step": 4140}, {"loss": 1.4348, "grad_norm": 0.6771699786186218, "learning_rate": 0.0002, "epoch": 3.220799379123011, "step": 4150}, {"loss": 1.4736, "grad_norm": 0.699035108089447, "learning_rate": 0.0002, "epoch": 3.2285603414823436, "step": 4160}, {"loss": 1.4096, "grad_norm": 0.7218514680862427, "learning_rate": 0.0002, "epoch": 3.2363213038416765, "step": 4170}, {"loss": 1.3637, "grad_norm": 0.6270631551742554, "learning_rate": 0.0002, "epoch": 3.244082266201009, "step": 4180}, {"loss": 1.4076, "grad_norm": 0.6828921437263489, "learning_rate": 0.0002, "epoch": 3.2518432285603414, "step": 4190}, {"loss": 1.4663, "grad_norm": 0.6005498170852661, "learning_rate": 0.0002, "epoch": 3.2596041909196742, "step": 4200}, {"loss": 1.4798, "grad_norm": 0.6974790692329407, "learning_rate": 0.0002, "epoch": 3.2673651532790067, "step": 4210}, {"loss": 1.5012, "grad_norm": 0.7269543409347534, "learning_rate": 0.0002, "epoch": 3.275126115638339, "step": 4220}, {"loss": 1.3848, "grad_norm": 0.6728787422180176, "learning_rate": 0.0002, "epoch": 3.2828870779976715, "step": 4230}, {"loss": 1.4112, "grad_norm": 0.676972508430481, "learning_rate": 0.0002, "epoch": 3.2906480403570044, "step": 4240}, {"loss": 1.4206, "grad_norm": 0.748309314250946, "learning_rate": 0.0002, "epoch": 3.298409002716337, "step": 4250}, {"loss": 1.4973, "grad_norm": 0.6976589560508728, "learning_rate": 0.0002, "epoch": 3.3061699650756693, "step": 4260}, {"loss": 1.3967, "grad_norm": 0.649780809879303, "learning_rate": 0.0002, "epoch": 3.3139309274350017, "step": 4270}, {"loss": 1.327, "grad_norm": 0.6529902815818787, "learning_rate": 0.0002, "epoch": 3.3216918897943346, "step": 4280}, {"loss": 1.4888, "grad_norm": 0.9273163676261902, "learning_rate": 0.0002, "epoch": 3.329452852153667, "step": 4290}, {"loss": 1.4859, "grad_norm": 0.717024028301239, "learning_rate": 0.0002, "epoch": 3.3372138145129995, "step": 4300}, {"loss": 1.4441, "grad_norm": 0.7914950251579285, "learning_rate": 0.0002, "epoch": 3.3449747768723324, "step": 4310}, {"loss": 1.432, "grad_norm": 0.7133203148841858, "learning_rate": 0.0002, "epoch": 3.352735739231665, "step": 4320}, {"loss": 1.4662, "grad_norm": 0.7409568428993225, "learning_rate": 0.0002, "epoch": 3.3604967015909972, "step": 4330}, {"loss": 1.3992, "grad_norm": 0.6993981003761292, "learning_rate": 0.0002, "epoch": 3.3682576639503297, "step": 4340}, {"loss": 1.4261, "grad_norm": 0.7114535570144653, "learning_rate": 0.0002, "epoch": 3.3760186263096625, "step": 4350}, {"loss": 1.4227, "grad_norm": 0.6790860295295715, "learning_rate": 0.0002, "epoch": 3.383779588668995, "step": 4360}, {"loss": 1.4128, "grad_norm": 0.6507849097251892, "learning_rate": 0.0002, "epoch": 3.3915405510283274, "step": 4370}, {"loss": 1.4559, "grad_norm": 0.5967804193496704, "learning_rate": 0.0002, "epoch": 3.39930151338766, "step": 4380}, {"loss": 1.3687, "grad_norm": 0.6625847816467285, "learning_rate": 0.0002, "epoch": 3.4070624757469927, "step": 4390}, {"loss": 1.4193, "grad_norm": 0.6736508011817932, "learning_rate": 0.0002, "epoch": 3.414823438106325, "step": 4400}, {"loss": 1.4363, "grad_norm": 0.7870860695838928, "learning_rate": 0.0002, "epoch": 3.4225844004656576, "step": 4410}, {"loss": 1.4114, "grad_norm": 0.7205295562744141, "learning_rate": 0.0002, "epoch": 3.4303453628249905, "step": 4420}, {"loss": 1.4131, "grad_norm": 0.6634634137153625, "learning_rate": 0.0002, "epoch": 3.438106325184323, "step": 4430}, {"loss": 1.4683, "grad_norm": 0.7562733292579651, "learning_rate": 0.0002, "epoch": 3.4458672875436553, "step": 4440}, {"loss": 1.3486, "grad_norm": 0.6585879921913147, "learning_rate": 0.0002, "epoch": 3.453628249902988, "step": 4450}, {"loss": 1.4283, "grad_norm": 0.6896792054176331, "learning_rate": 0.0002, "epoch": 3.4613892122623207, "step": 4460}, {"loss": 1.4208, "grad_norm": 0.6520342230796814, "learning_rate": 0.0002, "epoch": 3.469150174621653, "step": 4470}, {"loss": 1.3423, "grad_norm": 0.6760806441307068, "learning_rate": 0.0002, "epoch": 3.4769111369809855, "step": 4480}, {"loss": 1.4398, "grad_norm": 0.7539774179458618, "learning_rate": 0.0002, "epoch": 3.484672099340318, "step": 4490}, {"loss": 1.4534, "grad_norm": 0.7409411668777466, "learning_rate": 0.0002, "epoch": 3.492433061699651, "step": 4500}, {"loss": 1.4069, "grad_norm": 0.6876253485679626, "learning_rate": 0.0002, "epoch": 3.5001940240589833, "step": 4510}, {"loss": 1.4228, "grad_norm": 0.7028461694717407, "learning_rate": 0.0002, "epoch": 3.5079549864183157, "step": 4520}, {"loss": 1.4723, "grad_norm": 0.8056529760360718, "learning_rate": 0.0002, "epoch": 3.5157159487776486, "step": 4530}, {"loss": 1.4148, "grad_norm": 0.711338996887207, "learning_rate": 0.0002, "epoch": 3.523476911136981, "step": 4540}, {"loss": 1.5247, "grad_norm": 0.7343552708625793, "learning_rate": 0.0002, "epoch": 3.5312378734963135, "step": 4550}, {"loss": 1.4308, "grad_norm": 0.745479941368103, "learning_rate": 0.0002, "epoch": 3.5389988358556463, "step": 4560}, {"loss": 1.4229, "grad_norm": 0.7582294940948486, "learning_rate": 0.0002, "epoch": 3.5467597982149788, "step": 4570}, {"loss": 1.4127, "grad_norm": 0.6717444658279419, "learning_rate": 0.0002, "epoch": 3.554520760574311, "step": 4580}, {"loss": 1.4368, "grad_norm": 0.7417883276939392, "learning_rate": 0.0002, "epoch": 3.5622817229336436, "step": 4590}, {"loss": 1.4176, "grad_norm": 0.6385737061500549, "learning_rate": 0.0002, "epoch": 3.570042685292976, "step": 4600}, {"loss": 1.3981, "grad_norm": 0.716704249382019, "learning_rate": 0.0002, "epoch": 3.577803647652309, "step": 4610}, {"loss": 1.3889, "grad_norm": 0.6948980093002319, "learning_rate": 0.0002, "epoch": 3.5855646100116414, "step": 4620}, {"loss": 1.5177, "grad_norm": 0.6961140036582947, "learning_rate": 0.0002, "epoch": 3.593325572370974, "step": 4630}, {"loss": 1.4508, "grad_norm": 0.7493122220039368, "learning_rate": 0.0002, "epoch": 3.6010865347303067, "step": 4640}, {"loss": 1.3987, "grad_norm": 0.7431658506393433, "learning_rate": 0.0002, "epoch": 3.608847497089639, "step": 4650}, {"loss": 1.4551, "grad_norm": 0.8353387713432312, "learning_rate": 0.0002, "epoch": 3.6166084594489716, "step": 4660}, {"loss": 1.4533, "grad_norm": 0.7095612287521362, "learning_rate": 0.0002, "epoch": 3.6243694218083045, "step": 4670}, {"loss": 1.4003, "grad_norm": 0.776620090007782, "learning_rate": 0.0002, "epoch": 3.632130384167637, "step": 4680}, {"loss": 1.4361, "grad_norm": 0.7198925018310547, "learning_rate": 0.0002, "epoch": 3.6398913465269693, "step": 4690}, {"loss": 1.4543, "grad_norm": 0.8238834738731384, "learning_rate": 0.0002, "epoch": 3.6476523088863018, "step": 4700}, {"loss": 1.3958, "grad_norm": 0.6804245710372925, "learning_rate": 0.0002, "epoch": 3.655413271245634, "step": 4710}, {"loss": 1.4158, "grad_norm": 0.8444845676422119, "learning_rate": 0.0002, "epoch": 3.663174233604967, "step": 4720}, {"loss": 1.3825, "grad_norm": 0.743797779083252, "learning_rate": 0.0002, "epoch": 3.6709351959642995, "step": 4730}, {"loss": 1.4213, "grad_norm": 0.8994188904762268, "learning_rate": 0.0002, "epoch": 3.678696158323632, "step": 4740}, {"loss": 1.4281, "grad_norm": 0.75416100025177, "learning_rate": 0.0002, "epoch": 3.686457120682965, "step": 4750}, {"loss": 1.4154, "grad_norm": 0.6499266028404236, "learning_rate": 0.0002, "epoch": 3.6942180830422973, "step": 4760}, {"loss": 1.4005, "grad_norm": 0.7246791124343872, "learning_rate": 0.0002, "epoch": 3.7019790454016297, "step": 4770}, {"loss": 1.426, "grad_norm": 0.7831124067306519, "learning_rate": 0.0002, "epoch": 3.7097400077609626, "step": 4780}, {"loss": 1.3933, "grad_norm": 0.7130028009414673, "learning_rate": 0.0002, "epoch": 3.717500970120295, "step": 4790}, {"loss": 1.4632, "grad_norm": 0.7501602172851562, "learning_rate": 0.0002, "epoch": 3.7252619324796274, "step": 4800}, {"loss": 1.4985, "grad_norm": 0.6980932950973511, "learning_rate": 0.0002, "epoch": 3.73302289483896, "step": 4810}, {"loss": 1.4517, "grad_norm": 0.8050530552864075, "learning_rate": 0.0002, "epoch": 3.7407838571982923, "step": 4820}, {"loss": 1.4703, "grad_norm": 0.6385579705238342, "learning_rate": 0.0002, "epoch": 3.748544819557625, "step": 4830}, {"loss": 1.5281, "grad_norm": 0.6664714813232422, "learning_rate": 0.0002, "epoch": 3.7563057819169576, "step": 4840}, {"loss": 1.4443, "grad_norm": 0.7125676274299622, "learning_rate": 0.0002, "epoch": 3.76406674427629, "step": 4850}, {"loss": 1.3958, "grad_norm": 0.7231866717338562, "learning_rate": 0.0002, "epoch": 3.771827706635623, "step": 4860}, {"loss": 1.4446, "grad_norm": 0.6917183995246887, "learning_rate": 0.0002, "epoch": 3.7795886689949554, "step": 4870}, {"loss": 1.4369, "grad_norm": 0.665037989616394, "learning_rate": 0.0002, "epoch": 3.787349631354288, "step": 4880}, {"loss": 1.4193, "grad_norm": 0.5837726593017578, "learning_rate": 0.0002, "epoch": 3.7951105937136207, "step": 4890}, {"loss": 1.4176, "grad_norm": 0.6366701722145081, "learning_rate": 0.0002, "epoch": 3.802871556072953, "step": 4900}, {"loss": 1.46, "grad_norm": 0.7082223892211914, "learning_rate": 0.0002, "epoch": 3.8106325184322856, "step": 4910}, {"loss": 1.5139, "grad_norm": 0.8101672530174255, "learning_rate": 0.0002, "epoch": 3.818393480791618, "step": 4920}, {"loss": 1.3659, "grad_norm": 0.7516148090362549, "learning_rate": 0.0002, "epoch": 3.826154443150951, "step": 4930}, {"loss": 1.3909, "grad_norm": 0.7928489446640015, "learning_rate": 0.0002, "epoch": 3.8339154055102833, "step": 4940}, {"loss": 1.4255, "grad_norm": 0.6892234683036804, "learning_rate": 0.0002, "epoch": 3.8416763678696157, "step": 4950}, {"loss": 1.5024, "grad_norm": 0.6381304264068604, "learning_rate": 0.0002, "epoch": 3.849437330228948, "step": 4960}, {"loss": 1.4873, "grad_norm": 0.8068831562995911, "learning_rate": 0.0002, "epoch": 3.857198292588281, "step": 4970}, {"loss": 1.45, "grad_norm": 0.7289869785308838, "learning_rate": 0.0002, "epoch": 3.8649592549476135, "step": 4980}, {"loss": 1.398, "grad_norm": 0.7278549075126648, "learning_rate": 0.0002, "epoch": 3.872720217306946, "step": 4990}, {"loss": 1.4442, "grad_norm": 0.7324236631393433, "learning_rate": 0.0002, "epoch": 3.880481179666279, "step": 5000}, {"loss": 1.4511, "grad_norm": 0.6759871244430542, "learning_rate": 0.0002, "epoch": 3.8882421420256112, "step": 5010}, {"loss": 1.4705, "grad_norm": 0.8159207701683044, "learning_rate": 0.0002, "epoch": 3.8960031043849437, "step": 5020}, {"loss": 1.4685, "grad_norm": 0.6536211967468262, "learning_rate": 0.0002, "epoch": 3.9037640667442766, "step": 5030}, {"loss": 1.4335, "grad_norm": 0.6827932000160217, "learning_rate": 0.0002, "epoch": 3.911525029103609, "step": 5040}, {"loss": 1.433, "grad_norm": 0.6688340306282043, "learning_rate": 0.0002, "epoch": 3.9192859914629414, "step": 5050}, {"loss": 1.4099, "grad_norm": 0.6385695934295654, "learning_rate": 0.0002, "epoch": 3.927046953822274, "step": 5060}, {"loss": 1.4767, "grad_norm": 0.6975107192993164, "learning_rate": 0.0002, "epoch": 3.9348079161816063, "step": 5070}, {"loss": 1.4893, "grad_norm": 0.6684112548828125, "learning_rate": 0.0002, "epoch": 3.942568878540939, "step": 5080}, {"loss": 1.4732, "grad_norm": 0.8349628448486328, "learning_rate": 0.0002, "epoch": 3.9503298409002716, "step": 5090}, {"loss": 1.5131, "grad_norm": 0.7146425843238831, "learning_rate": 0.0002, "epoch": 3.958090803259604, "step": 5100}, {"loss": 1.4149, "grad_norm": 0.6555036902427673, "learning_rate": 0.0002, "epoch": 3.965851765618937, "step": 5110}, {"loss": 1.4274, "grad_norm": 0.7037415504455566, "learning_rate": 0.0002, "epoch": 3.9736127279782694, "step": 5120}, {"loss": 1.4292, "grad_norm": 0.7235575914382935, "learning_rate": 0.0002, "epoch": 3.981373690337602, "step": 5130}, {"loss": 1.4455, "grad_norm": 0.7092325687408447, "learning_rate": 0.0002, "epoch": 3.9891346526969347, "step": 5140}, {"loss": 1.4512, "grad_norm": 0.7490319609642029, "learning_rate": 0.0002, "epoch": 3.996895615056267, "step": 5150}, {"eval_loss": 1.9131355285644531, "eval_runtime": 105.5778, "eval_samples_per_second": 4.802, "eval_steps_per_second": 0.606, "epoch": 4.0, "step": 5154}, {"loss": 1.2643, "grad_norm": 0.7075854539871216, "learning_rate": 0.0002, "epoch": 4.0046565774155995, "step": 5160}, {"loss": 1.209, "grad_norm": 0.9466007351875305, "learning_rate": 0.0002, "epoch": 4.012417539774932, "step": 5170}, {"loss": 1.2567, "grad_norm": 1.0297044515609741, "learning_rate": 0.0002, "epoch": 4.020178502134264, "step": 5180}, {"loss": 1.1796, "grad_norm": 0.7765059471130371, "learning_rate": 0.0002, "epoch": 4.027939464493597, "step": 5190}, {"loss": 1.2356, "grad_norm": 0.995760977268219, "learning_rate": 0.0002, "epoch": 4.03570042685293, "step": 5200}, {"loss": 1.1792, "grad_norm": 0.8663829565048218, "learning_rate": 0.0002, "epoch": 4.043461389212262, "step": 5210}, {"loss": 1.2471, "grad_norm": 1.0660825967788696, "learning_rate": 0.0002, "epoch": 4.051222351571595, "step": 5220}, {"loss": 1.1676, "grad_norm": 0.9858174920082092, "learning_rate": 0.0002, "epoch": 4.058983313930927, "step": 5230}, {"loss": 1.2448, "grad_norm": 0.8911338448524475, "learning_rate": 0.0002, "epoch": 4.06674427629026, "step": 5240}, {"loss": 1.1858, "grad_norm": 1.0848394632339478, "learning_rate": 0.0002, "epoch": 4.074505238649593, "step": 5250}, {"loss": 1.1684, "grad_norm": 1.0849905014038086, "learning_rate": 0.0002, "epoch": 4.082266201008925, "step": 5260}, {"loss": 1.2007, "grad_norm": 1.0497841835021973, "learning_rate": 0.0002, "epoch": 4.090027163368258, "step": 5270}, {"loss": 1.2552, "grad_norm": 0.8943053483963013, "learning_rate": 0.0002, "epoch": 4.0977881257275905, "step": 5280}, {"loss": 1.1923, "grad_norm": 0.8432527184486389, "learning_rate": 0.0002, "epoch": 4.1055490880869225, "step": 5290}, {"loss": 1.1634, "grad_norm": 0.9690414667129517, "learning_rate": 0.0002, "epoch": 4.113310050446255, "step": 5300}, {"loss": 1.3019, "grad_norm": 0.7790773510932922, "learning_rate": 0.0002, "epoch": 4.121071012805588, "step": 5310}, {"loss": 1.1806, "grad_norm": 0.9289211630821228, "learning_rate": 0.0002, "epoch": 4.12883197516492, "step": 5320}, {"loss": 1.1458, "grad_norm": 1.0785125494003296, "learning_rate": 0.0002, "epoch": 4.136592937524253, "step": 5330}, {"loss": 1.2086, "grad_norm": 0.8559591770172119, "learning_rate": 0.0002, "epoch": 4.144353899883585, "step": 5340}, {"loss": 1.1974, "grad_norm": 0.9405956268310547, "learning_rate": 0.0002, "epoch": 4.152114862242918, "step": 5350}, {"loss": 1.1793, "grad_norm": 0.9942827820777893, "learning_rate": 0.0002, "epoch": 4.159875824602251, "step": 5360}, {"loss": 1.1659, "grad_norm": 0.9141933917999268, "learning_rate": 0.0002, "epoch": 4.167636786961583, "step": 5370}, {"loss": 1.1647, "grad_norm": 0.8206015229225159, "learning_rate": 0.0002, "epoch": 4.175397749320916, "step": 5380}, {"loss": 1.2778, "grad_norm": 0.9340888857841492, "learning_rate": 0.0002, "epoch": 4.183158711680249, "step": 5390}, {"loss": 1.2459, "grad_norm": 1.2122114896774292, "learning_rate": 0.0002, "epoch": 4.190919674039581, "step": 5400}, {"loss": 1.2371, "grad_norm": 1.0661298036575317, "learning_rate": 0.0002, "epoch": 4.1986806363989135, "step": 5410}, {"loss": 1.1978, "grad_norm": 0.9372861385345459, "learning_rate": 0.0002, "epoch": 4.206441598758246, "step": 5420}, {"loss": 1.2653, "grad_norm": 0.894012987613678, "learning_rate": 0.0002, "epoch": 4.214202561117578, "step": 5430}, {"loss": 1.387, "grad_norm": 1.0647753477096558, "learning_rate": 0.0002, "epoch": 4.221963523476911, "step": 5440}, {"loss": 1.2231, "grad_norm": 0.989179790019989, "learning_rate": 0.0002, "epoch": 4.229724485836243, "step": 5450}, {"loss": 1.2715, "grad_norm": 1.1601181030273438, "learning_rate": 0.0002, "epoch": 4.237485448195576, "step": 5460}, {"loss": 1.2406, "grad_norm": 0.9395585656166077, "learning_rate": 0.0002, "epoch": 4.245246410554909, "step": 5470}, {"loss": 1.2779, "grad_norm": 0.9527766108512878, "learning_rate": 0.0002, "epoch": 4.253007372914241, "step": 5480}, {"loss": 1.267, "grad_norm": 1.0319520235061646, "learning_rate": 0.0002, "epoch": 4.260768335273574, "step": 5490}, {"loss": 1.2633, "grad_norm": 0.8659824728965759, "learning_rate": 0.0002, "epoch": 4.268529297632907, "step": 5500}, {"loss": 1.1475, "grad_norm": 1.099211573600769, "learning_rate": 0.0002, "epoch": 4.276290259992239, "step": 5510}, {"loss": 1.2508, "grad_norm": 0.9363361597061157, "learning_rate": 0.0002, "epoch": 4.284051222351572, "step": 5520}, {"loss": 1.189, "grad_norm": 0.8437647223472595, "learning_rate": 0.0002, "epoch": 4.2918121847109045, "step": 5530}, {"loss": 1.2212, "grad_norm": 0.9181258678436279, "learning_rate": 0.0002, "epoch": 4.2995731470702365, "step": 5540}, {"loss": 1.2092, "grad_norm": 0.9059357643127441, "learning_rate": 0.0002, "epoch": 4.307334109429569, "step": 5550}, {"loss": 1.2189, "grad_norm": 0.9337241649627686, "learning_rate": 0.0002, "epoch": 4.315095071788901, "step": 5560}, {"loss": 1.2462, "grad_norm": 0.9428889155387878, "learning_rate": 0.0002, "epoch": 4.322856034148234, "step": 5570}, {"loss": 1.2675, "grad_norm": 1.003589153289795, "learning_rate": 0.0002, "epoch": 4.330616996507567, "step": 5580}, {"loss": 1.2703, "grad_norm": 1.1249268054962158, "learning_rate": 0.0002, "epoch": 4.338377958866899, "step": 5590}, {"loss": 1.2501, "grad_norm": 0.8623469471931458, "learning_rate": 0.0002, "epoch": 4.346138921226232, "step": 5600}, {"loss": 1.2404, "grad_norm": 1.1389174461364746, "learning_rate": 0.0002, "epoch": 4.353899883585565, "step": 5610}, {"loss": 1.2245, "grad_norm": 1.0136264562606812, "learning_rate": 0.0002, "epoch": 4.361660845944897, "step": 5620}, {"loss": 1.3473, "grad_norm": 0.9567070603370667, "learning_rate": 0.0002, "epoch": 4.36942180830423, "step": 5630}, {"loss": 1.2988, "grad_norm": 1.0592148303985596, "learning_rate": 0.0002, "epoch": 4.377182770663563, "step": 5640}, {"loss": 1.212, "grad_norm": 1.0110485553741455, "learning_rate": 0.0002, "epoch": 4.384943733022895, "step": 5650}, {"loss": 1.2086, "grad_norm": 0.9914907217025757, "learning_rate": 0.0002, "epoch": 4.3927046953822275, "step": 5660}, {"loss": 1.2363, "grad_norm": 0.9447247982025146, "learning_rate": 0.0002, "epoch": 4.4004656577415595, "step": 5670}, {"loss": 1.2617, "grad_norm": 0.9644378423690796, "learning_rate": 0.0002, "epoch": 4.408226620100892, "step": 5680}, {"loss": 1.2773, "grad_norm": 0.920676589012146, "learning_rate": 0.0002, "epoch": 4.415987582460225, "step": 5690}, {"loss": 1.2792, "grad_norm": 1.060570478439331, "learning_rate": 0.0002, "epoch": 4.423748544819557, "step": 5700}, {"loss": 1.2374, "grad_norm": 0.8857738971710205, "learning_rate": 0.0002, "epoch": 4.43150950717889, "step": 5710}, {"loss": 1.2588, "grad_norm": 1.0536398887634277, "learning_rate": 0.0002, "epoch": 4.439270469538223, "step": 5720}, {"loss": 1.2051, "grad_norm": 0.990847110748291, "learning_rate": 0.0002, "epoch": 4.447031431897555, "step": 5730}, {"loss": 1.2469, "grad_norm": 0.9692499041557312, "learning_rate": 0.0002, "epoch": 4.454792394256888, "step": 5740}, {"loss": 1.2269, "grad_norm": 1.0376402139663696, "learning_rate": 0.0002, "epoch": 4.462553356616221, "step": 5750}, {"loss": 1.1701, "grad_norm": 1.3863259553909302, "learning_rate": 0.0002, "epoch": 4.470314318975553, "step": 5760}, {"loss": 1.2591, "grad_norm": 0.978379487991333, "learning_rate": 0.0002, "epoch": 4.478075281334886, "step": 5770}, {"loss": 1.2729, "grad_norm": 1.0973085165023804, "learning_rate": 0.0002, "epoch": 4.485836243694218, "step": 5780}, {"loss": 1.2404, "grad_norm": 1.057006597518921, "learning_rate": 0.0002, "epoch": 4.4935972060535505, "step": 5790}, {"loss": 1.2476, "grad_norm": 0.9247729182243347, "learning_rate": 0.0002, "epoch": 4.501358168412883, "step": 5800}, {"loss": 1.2369, "grad_norm": 1.0447787046432495, "learning_rate": 0.0002, "epoch": 4.509119130772215, "step": 5810}, {"loss": 1.211, "grad_norm": 1.1930429935455322, "learning_rate": 0.0002, "epoch": 4.516880093131548, "step": 5820}, {"loss": 1.2596, "grad_norm": 0.9867590069770813, "learning_rate": 0.0002, "epoch": 4.524641055490881, "step": 5830}, {"loss": 1.2766, "grad_norm": 0.9591100215911865, "learning_rate": 0.0002, "epoch": 4.532402017850213, "step": 5840}, {"loss": 1.2154, "grad_norm": 0.9950753450393677, "learning_rate": 0.0002, "epoch": 4.540162980209546, "step": 5850}, {"loss": 1.2149, "grad_norm": 1.0087506771087646, "learning_rate": 0.0002, "epoch": 4.547923942568879, "step": 5860}, {"loss": 1.3165, "grad_norm": 1.0934417247772217, "learning_rate": 0.0002, "epoch": 4.555684904928211, "step": 5870}, {"loss": 1.3059, "grad_norm": 1.107987403869629, "learning_rate": 0.0002, "epoch": 4.563445867287544, "step": 5880}, {"loss": 1.2184, "grad_norm": 0.9147276878356934, "learning_rate": 0.0002, "epoch": 4.571206829646876, "step": 5890}, {"loss": 1.24, "grad_norm": 1.036780595779419, "learning_rate": 0.0002, "epoch": 4.578967792006209, "step": 5900}, {"loss": 1.2209, "grad_norm": 0.9284719824790955, "learning_rate": 0.0002, "epoch": 4.5867287543655415, "step": 5910}, {"loss": 1.3693, "grad_norm": 0.9141898155212402, "learning_rate": 0.0002, "epoch": 4.5944897167248735, "step": 5920}, {"loss": 1.2319, "grad_norm": 1.0447357892990112, "learning_rate": 0.0002, "epoch": 4.602250679084206, "step": 5930}, {"loss": 1.2667, "grad_norm": 0.9309114217758179, "learning_rate": 0.0002, "epoch": 4.610011641443539, "step": 5940}, {"loss": 1.2827, "grad_norm": 1.2986129522323608, "learning_rate": 0.0002, "epoch": 4.617772603802871, "step": 5950}, {"loss": 1.312, "grad_norm": 0.9221704602241516, "learning_rate": 0.0002, "epoch": 4.625533566162204, "step": 5960}, {"loss": 1.2769, "grad_norm": 0.9228187799453735, "learning_rate": 0.0002, "epoch": 4.633294528521537, "step": 5970}, {"loss": 1.2953, "grad_norm": 0.9483116269111633, "learning_rate": 0.0002, "epoch": 4.641055490880869, "step": 5980}, {"loss": 1.3437, "grad_norm": 1.0218974351882935, "learning_rate": 0.0002, "epoch": 4.648816453240202, "step": 5990}, {"loss": 1.3085, "grad_norm": 0.9764600396156311, "learning_rate": 0.0002, "epoch": 4.656577415599534, "step": 6000}, {"loss": 1.197, "grad_norm": 0.9115710258483887, "learning_rate": 0.0002, "epoch": 4.664338377958867, "step": 6010}, {"loss": 1.1917, "grad_norm": 0.9245651364326477, "learning_rate": 0.0002, "epoch": 4.6720993403182, "step": 6020}, {"loss": 1.2969, "grad_norm": 0.9686311483383179, "learning_rate": 0.0002, "epoch": 4.6798603026775325, "step": 6030}, {"loss": 1.2702, "grad_norm": 1.1807392835617065, "learning_rate": 0.0002, "epoch": 4.6876212650368645, "step": 6040}, {"loss": 1.328, "grad_norm": 1.0358641147613525, "learning_rate": 0.0002, "epoch": 4.695382227396197, "step": 6050}, {"loss": 1.3281, "grad_norm": 0.987332284450531, "learning_rate": 0.0002, "epoch": 4.703143189755529, "step": 6060}, {"loss": 1.2514, "grad_norm": 1.0526494979858398, "learning_rate": 0.0002, "epoch": 4.710904152114862, "step": 6070}, {"loss": 1.2246, "grad_norm": 1.0276758670806885, "learning_rate": 0.0002, "epoch": 4.718665114474195, "step": 6080}, {"loss": 1.3367, "grad_norm": 0.9904406666755676, "learning_rate": 0.0002, "epoch": 4.726426076833527, "step": 6090}, {"loss": 1.2797, "grad_norm": 1.0084882974624634, "learning_rate": 0.0002, "epoch": 4.73418703919286, "step": 6100}, {"loss": 1.2656, "grad_norm": 0.8646450638771057, "learning_rate": 0.0002, "epoch": 4.741948001552192, "step": 6110}, {"loss": 1.3063, "grad_norm": 0.9233377575874329, "learning_rate": 0.0002, "epoch": 4.749708963911525, "step": 6120}, {"loss": 1.2642, "grad_norm": 0.9675140976905823, "learning_rate": 0.0002, "epoch": 4.757469926270858, "step": 6130}, {"loss": 1.3367, "grad_norm": 0.9639796018600464, "learning_rate": 0.0002, "epoch": 4.765230888630191, "step": 6140}, {"loss": 1.276, "grad_norm": 0.925199568271637, "learning_rate": 0.0002, "epoch": 4.772991850989523, "step": 6150}, {"loss": 1.2441, "grad_norm": 1.050901174545288, "learning_rate": 0.0002, "epoch": 4.7807528133488555, "step": 6160}, {"loss": 1.301, "grad_norm": 0.8920623660087585, "learning_rate": 0.0002, "epoch": 4.7885137757081875, "step": 6170}, {"loss": 1.263, "grad_norm": 0.8964757919311523, "learning_rate": 0.0002, "epoch": 4.79627473806752, "step": 6180}, {"loss": 1.2787, "grad_norm": 1.0839070081710815, "learning_rate": 0.0002, "epoch": 4.804035700426853, "step": 6190}, {"loss": 1.2664, "grad_norm": 0.8809942007064819, "learning_rate": 0.0002, "epoch": 4.811796662786185, "step": 6200}, {"loss": 1.321, "grad_norm": 1.0216195583343506, "learning_rate": 0.0002, "epoch": 4.819557625145518, "step": 6210}, {"loss": 1.3033, "grad_norm": 0.892005980014801, "learning_rate": 0.0002, "epoch": 4.827318587504851, "step": 6220}, {"loss": 1.2602, "grad_norm": 0.9957166910171509, "learning_rate": 0.0002, "epoch": 4.835079549864183, "step": 6230}, {"loss": 1.3562, "grad_norm": 0.9720533490180969, "learning_rate": 0.0002, "epoch": 4.842840512223516, "step": 6240}, {"loss": 1.2651, "grad_norm": 0.9336182475090027, "learning_rate": 0.0002, "epoch": 4.850601474582849, "step": 6250}, {"loss": 1.3136, "grad_norm": 1.2611457109451294, "learning_rate": 0.0002, "epoch": 4.858362436942181, "step": 6260}, {"loss": 1.2234, "grad_norm": 0.8927203416824341, "learning_rate": 0.0002, "epoch": 4.866123399301514, "step": 6270}, {"loss": 1.3463, "grad_norm": 0.9706710577011108, "learning_rate": 0.0002, "epoch": 4.873884361660846, "step": 6280}, {"loss": 1.3209, "grad_norm": 1.1461690664291382, "learning_rate": 0.0002, "epoch": 4.8816453240201785, "step": 6290}, {"loss": 1.2566, "grad_norm": 0.9930381178855896, "learning_rate": 0.0002, "epoch": 4.889406286379511, "step": 6300}, {"loss": 1.2568, "grad_norm": 0.91451096534729, "learning_rate": 0.0002, "epoch": 4.897167248738843, "step": 6310}, {"loss": 1.2836, "grad_norm": 1.0319571495056152, "learning_rate": 0.0002, "epoch": 4.904928211098176, "step": 6320}, {"loss": 1.2908, "grad_norm": 0.990140438079834, "learning_rate": 0.0002, "epoch": 4.912689173457509, "step": 6330}, {"loss": 1.3299, "grad_norm": 1.2466117143630981, "learning_rate": 0.0002, "epoch": 4.920450135816841, "step": 6340}, {"loss": 1.2659, "grad_norm": 1.0316979885101318, "learning_rate": 0.0002, "epoch": 4.928211098176174, "step": 6350}, {"loss": 1.3292, "grad_norm": 1.0643759965896606, "learning_rate": 0.0002, "epoch": 4.935972060535507, "step": 6360}, {"loss": 1.2559, "grad_norm": 0.9703279733657837, "learning_rate": 0.0002, "epoch": 4.943733022894839, "step": 6370}, {"loss": 1.2155, "grad_norm": 0.9767927527427673, "learning_rate": 0.0002, "epoch": 4.951493985254172, "step": 6380}, {"loss": 1.2437, "grad_norm": 0.960854172706604, "learning_rate": 0.0002, "epoch": 4.959254947613504, "step": 6390}, {"loss": 1.3314, "grad_norm": 0.9922910332679749, "learning_rate": 0.0002, "epoch": 4.967015909972837, "step": 6400}, {"loss": 1.3018, "grad_norm": 0.956470787525177, "learning_rate": 0.0002, "epoch": 4.9747768723321695, "step": 6410}, {"loss": 1.2794, "grad_norm": 0.9637242555618286, "learning_rate": 0.0002, "epoch": 4.9825378346915015, "step": 6420}, {"loss": 1.3236, "grad_norm": 1.0855202674865723, "learning_rate": 0.0002, "epoch": 4.990298797050834, "step": 6430}, {"loss": 1.3015, "grad_norm": 0.9655316472053528, "learning_rate": 0.0002, "epoch": 4.998059759410167, "step": 6440}, {"eval_loss": 2.0410802364349365, "eval_runtime": 113.04, "eval_samples_per_second": 4.485, "eval_steps_per_second": 0.566, "epoch": 4.9996119518820334, "step": 6442}, {"loss": 1.0846, "grad_norm": 1.1676199436187744, "learning_rate": 0.0002, "epoch": 5.005820721769499, "step": 6450}, {"loss": 1.041, "grad_norm": 1.4317965507507324, "learning_rate": 0.0002, "epoch": 5.013581684128832, "step": 6460}, {"loss": 0.9546, "grad_norm": 1.460443377494812, "learning_rate": 0.0002, "epoch": 5.021342646488165, "step": 6470}, {"loss": 1.0014, "grad_norm": 1.2299214601516724, "learning_rate": 0.0002, "epoch": 5.029103608847497, "step": 6480}, {"loss": 1.0397, "grad_norm": 1.3125724792480469, "learning_rate": 0.0002, "epoch": 5.03686457120683, "step": 6490}, {"loss": 1.0134, "grad_norm": 1.1252319812774658, "learning_rate": 0.0002, "epoch": 5.044625533566162, "step": 6500}, {"loss": 0.976, "grad_norm": 0.9970866441726685, "learning_rate": 0.0002, "epoch": 5.052386495925495, "step": 6510}, {"loss": 0.9731, "grad_norm": 1.229069709777832, "learning_rate": 0.0002, "epoch": 5.060147458284828, "step": 6520}, {"loss": 1.0498, "grad_norm": 1.2430938482284546, "learning_rate": 0.0002, "epoch": 5.06790842064416, "step": 6530}, {"loss": 1.0236, "grad_norm": 1.0522737503051758, "learning_rate": 0.0002, "epoch": 5.0756693830034925, "step": 6540}, {"loss": 1.0221, "grad_norm": 1.108890175819397, "learning_rate": 0.0002, "epoch": 5.083430345362825, "step": 6550}, {"loss": 1.0177, "grad_norm": 1.156912922859192, "learning_rate": 0.0002, "epoch": 5.091191307722157, "step": 6560}, {"loss": 1.0415, "grad_norm": 1.405895709991455, "learning_rate": 0.0002, "epoch": 5.09895227008149, "step": 6570}, {"loss": 0.9811, "grad_norm": 1.2005155086517334, "learning_rate": 0.0002, "epoch": 5.106713232440823, "step": 6580}, {"loss": 0.9862, "grad_norm": 1.181443452835083, "learning_rate": 0.0002, "epoch": 5.114474194800155, "step": 6590}, {"loss": 1.0291, "grad_norm": 2.3444771766662598, "learning_rate": 0.0002, "epoch": 5.122235157159488, "step": 6600}, {"loss": 1.0455, "grad_norm": 1.216988444328308, "learning_rate": 0.0002, "epoch": 5.12999611951882, "step": 6610}, {"loss": 1.0549, "grad_norm": 1.369553565979004, "learning_rate": 0.0002, "epoch": 5.137757081878153, "step": 6620}, {"loss": 1.0056, "grad_norm": 1.177964687347412, "learning_rate": 0.0002, "epoch": 5.145518044237486, "step": 6630}, {"loss": 1.1025, "grad_norm": 1.1397041082382202, "learning_rate": 0.0002, "epoch": 5.153279006596818, "step": 6640}, {"loss": 1.0437, "grad_norm": 1.3976861238479614, "learning_rate": 0.0002, "epoch": 5.161039968956151, "step": 6650}, {"loss": 1.0454, "grad_norm": 1.4824495315551758, "learning_rate": 0.0002, "epoch": 5.1688009313154835, "step": 6660}, {"loss": 1.0356, "grad_norm": 1.2653018236160278, "learning_rate": 0.0002, "epoch": 5.1765618936748155, "step": 6670}, {"loss": 0.9971, "grad_norm": 1.3106069564819336, "learning_rate": 0.0002, "epoch": 5.184322856034148, "step": 6680}, {"loss": 1.0561, "grad_norm": 1.3140279054641724, "learning_rate": 0.0002, "epoch": 5.192083818393481, "step": 6690}, {"loss": 1.0618, "grad_norm": 1.3900256156921387, "learning_rate": 0.0002, "epoch": 5.199844780752813, "step": 6700}, {"loss": 1.0285, "grad_norm": 1.3191124200820923, "learning_rate": 0.0002, "epoch": 5.207605743112146, "step": 6710}, {"loss": 0.9921, "grad_norm": 1.176107406616211, "learning_rate": 0.0002, "epoch": 5.215366705471478, "step": 6720}, {"loss": 1.064, "grad_norm": 1.2364883422851562, "learning_rate": 0.0002, "epoch": 5.223127667830811, "step": 6730}, {"loss": 0.9599, "grad_norm": 1.343022108078003, "learning_rate": 0.0002, "epoch": 5.230888630190144, "step": 6740}, {"loss": 1.0342, "grad_norm": 1.2826898097991943, "learning_rate": 0.0002, "epoch": 5.238649592549476, "step": 6750}, {"loss": 1.0703, "grad_norm": 1.500257134437561, "learning_rate": 0.0002, "epoch": 5.246410554908809, "step": 6760}, {"loss": 1.0114, "grad_norm": 1.2605743408203125, "learning_rate": 0.0002, "epoch": 5.254171517268142, "step": 6770}, {"loss": 1.0825, "grad_norm": 1.2355525493621826, "learning_rate": 0.0002, "epoch": 5.261932479627474, "step": 6780}, {"loss": 1.0436, "grad_norm": 1.2845789194107056, "learning_rate": 0.0002, "epoch": 5.2696934419868064, "step": 6790}, {"loss": 0.989, "grad_norm": 1.3696625232696533, "learning_rate": 0.0002, "epoch": 5.277454404346139, "step": 6800}, {"loss": 1.0991, "grad_norm": 1.4051260948181152, "learning_rate": 0.0002, "epoch": 5.285215366705471, "step": 6810}, {"loss": 1.0987, "grad_norm": 1.266725778579712, "learning_rate": 0.0002, "epoch": 5.292976329064804, "step": 6820}, {"loss": 1.0489, "grad_norm": 1.3475236892700195, "learning_rate": 0.0002, "epoch": 5.300737291424136, "step": 6830}, {"loss": 1.0264, "grad_norm": 1.54409921169281, "learning_rate": 0.0002, "epoch": 5.308498253783469, "step": 6840}, {"loss": 1.033, "grad_norm": 1.2391985654830933, "learning_rate": 0.0002, "epoch": 5.316259216142802, "step": 6850}, {"loss": 1.1058, "grad_norm": 1.2435699701309204, "learning_rate": 0.0002, "epoch": 5.324020178502134, "step": 6860}, {"loss": 1.0179, "grad_norm": 1.8803037405014038, "learning_rate": 0.0002, "epoch": 5.331781140861467, "step": 6870}, {"loss": 0.997, "grad_norm": 1.4195542335510254, "learning_rate": 0.0002, "epoch": 5.3395421032208, "step": 6880}, {"loss": 1.0273, "grad_norm": 1.1853394508361816, "learning_rate": 0.0002, "epoch": 5.347303065580132, "step": 6890}, {"loss": 1.0668, "grad_norm": 1.4016530513763428, "learning_rate": 0.0002, "epoch": 5.355064027939465, "step": 6900}, {"loss": 1.1099, "grad_norm": 1.294339895248413, "learning_rate": 0.0002, "epoch": 5.3628249902987974, "step": 6910}, {"loss": 1.0724, "grad_norm": 1.2952708005905151, "learning_rate": 0.0002, "epoch": 5.370585952658129, "step": 6920}, {"loss": 1.0098, "grad_norm": 1.1361510753631592, "learning_rate": 0.0002, "epoch": 5.378346915017462, "step": 6930}, {"loss": 1.0796, "grad_norm": 1.125805377960205, "learning_rate": 0.0002, "epoch": 5.386107877376794, "step": 6940}, {"loss": 1.122, "grad_norm": 1.1453300714492798, "learning_rate": 0.0002, "epoch": 5.393868839736127, "step": 6950}, {"loss": 1.0977, "grad_norm": 1.4542768001556396, "learning_rate": 0.0002, "epoch": 5.40162980209546, "step": 6960}, {"loss": 1.0825, "grad_norm": 1.2360988855361938, "learning_rate": 0.0002, "epoch": 5.409390764454792, "step": 6970}, {"loss": 1.0631, "grad_norm": 1.2182754278182983, "learning_rate": 0.0002, "epoch": 5.417151726814125, "step": 6980}, {"loss": 1.0471, "grad_norm": 1.2018693685531616, "learning_rate": 0.0002, "epoch": 5.424912689173458, "step": 6990}, {"loss": 1.108, "grad_norm": 1.346124291419983, "learning_rate": 0.0002, "epoch": 5.43267365153279, "step": 7000}, {"loss": 1.0534, "grad_norm": 1.2534189224243164, "learning_rate": 0.0002, "epoch": 5.440434613892123, "step": 7010}, {"loss": 1.0696, "grad_norm": 1.2033339738845825, "learning_rate": 0.0002, "epoch": 5.448195576251456, "step": 7020}, {"loss": 1.0714, "grad_norm": 1.2788134813308716, "learning_rate": 0.0002, "epoch": 5.4559565386107876, "step": 7030}, {"loss": 1.1274, "grad_norm": 1.2751542329788208, "learning_rate": 0.0002, "epoch": 5.46371750097012, "step": 7040}, {"loss": 1.0767, "grad_norm": 1.3237019777297974, "learning_rate": 0.0002, "epoch": 5.471478463329452, "step": 7050}, {"loss": 1.1081, "grad_norm": 1.4932852983474731, "learning_rate": 0.0002, "epoch": 5.479239425688785, "step": 7060}, {"loss": 1.0197, "grad_norm": 1.4003876447677612, "learning_rate": 0.0002, "epoch": 5.487000388048118, "step": 7070}, {"loss": 1.0662, "grad_norm": 1.404799461364746, "learning_rate": 0.0002, "epoch": 5.49476135040745, "step": 7080}, {"loss": 1.0354, "grad_norm": 1.4486982822418213, "learning_rate": 0.0002, "epoch": 5.502522312766783, "step": 7090}, {"loss": 1.0645, "grad_norm": 1.1713480949401855, "learning_rate": 0.0002, "epoch": 5.510283275126116, "step": 7100}, {"loss": 1.006, "grad_norm": 1.4062601327896118, "learning_rate": 0.0002, "epoch": 5.518044237485448, "step": 7110}, {"loss": 1.0459, "grad_norm": 1.211629867553711, "learning_rate": 0.0002, "epoch": 5.525805199844781, "step": 7120}, {"loss": 1.102, "grad_norm": 1.2523176670074463, "learning_rate": 0.0002, "epoch": 5.533566162204114, "step": 7130}, {"loss": 1.1132, "grad_norm": 1.4467198848724365, "learning_rate": 0.0002, "epoch": 5.541327124563446, "step": 7140}, {"loss": 1.1557, "grad_norm": 1.5961614847183228, "learning_rate": 0.0002, "epoch": 5.5490880869227786, "step": 7150}, {"loss": 1.0859, "grad_norm": 1.320656418800354, "learning_rate": 0.0002, "epoch": 5.5568490492821105, "step": 7160}, {"loss": 1.109, "grad_norm": 1.2423332929611206, "learning_rate": 0.0002, "epoch": 5.564610011641443, "step": 7170}, {"loss": 1.0046, "grad_norm": 1.2919669151306152, "learning_rate": 0.0002, "epoch": 5.572370974000776, "step": 7180}, {"loss": 1.046, "grad_norm": 1.1678385734558105, "learning_rate": 0.0002, "epoch": 5.580131936360108, "step": 7190}, {"loss": 1.1011, "grad_norm": 1.4250764846801758, "learning_rate": 0.0002, "epoch": 5.587892898719441, "step": 7200}, {"loss": 1.1254, "grad_norm": 1.5308716297149658, "learning_rate": 0.0002, "epoch": 5.595653861078774, "step": 7210}, {"loss": 1.121, "grad_norm": 1.2678815126419067, "learning_rate": 0.0002, "epoch": 5.603414823438106, "step": 7220}, {"loss": 1.0846, "grad_norm": 1.127856969833374, "learning_rate": 0.0002, "epoch": 5.611175785797439, "step": 7230}, {"loss": 1.0647, "grad_norm": 1.3832560777664185, "learning_rate": 0.0002, "epoch": 5.618936748156772, "step": 7240}, {"loss": 1.0658, "grad_norm": 1.3226919174194336, "learning_rate": 0.0002, "epoch": 5.626697710516104, "step": 7250}, {"loss": 1.1175, "grad_norm": 1.3418006896972656, "learning_rate": 0.0002, "epoch": 5.634458672875437, "step": 7260}, {"loss": 1.0956, "grad_norm": 1.2625300884246826, "learning_rate": 0.0002, "epoch": 5.642219635234769, "step": 7270}, {"loss": 1.067, "grad_norm": 1.1579464673995972, "learning_rate": 0.0002, "epoch": 5.6499805975941015, "step": 7280}, {"loss": 1.0447, "grad_norm": 1.4998650550842285, "learning_rate": 0.0002, "epoch": 5.657741559953434, "step": 7290}, {"loss": 1.1256, "grad_norm": 1.2670758962631226, "learning_rate": 0.0002, "epoch": 5.665502522312766, "step": 7300}, {"loss": 1.1267, "grad_norm": 1.2959760427474976, "learning_rate": 0.0002, "epoch": 5.673263484672099, "step": 7310}, {"loss": 1.1387, "grad_norm": 1.2460671663284302, "learning_rate": 0.0002, "epoch": 5.681024447031432, "step": 7320}, {"loss": 1.0756, "grad_norm": 1.1313989162445068, "learning_rate": 0.0002, "epoch": 5.688785409390764, "step": 7330}, {"loss": 1.0618, "grad_norm": 1.282527208328247, "learning_rate": 0.0002, "epoch": 5.696546371750097, "step": 7340}, {"loss": 1.1315, "grad_norm": 1.3380206823349, "learning_rate": 0.0002, "epoch": 5.70430733410943, "step": 7350}, {"loss": 1.0949, "grad_norm": 1.1648279428482056, "learning_rate": 0.0002, "epoch": 5.712068296468762, "step": 7360}, {"loss": 1.1705, "grad_norm": 1.3059816360473633, "learning_rate": 0.0002, "epoch": 5.719829258828095, "step": 7370}, {"loss": 1.1496, "grad_norm": 1.1905046701431274, "learning_rate": 0.0002, "epoch": 5.727590221187427, "step": 7380}, {"loss": 1.1356, "grad_norm": 1.4089630842208862, "learning_rate": 0.0002, "epoch": 5.73535118354676, "step": 7390}, {"loss": 1.1349, "grad_norm": 1.256721019744873, "learning_rate": 0.0002, "epoch": 5.7431121459060925, "step": 7400}, {"loss": 1.0682, "grad_norm": 1.1915162801742554, "learning_rate": 0.0002, "epoch": 5.7508731082654245, "step": 7410}, {"loss": 1.1257, "grad_norm": 1.1935480833053589, "learning_rate": 0.0002, "epoch": 5.758634070624757, "step": 7420}, {"loss": 1.1348, "grad_norm": 1.1761008501052856, "learning_rate": 0.0002, "epoch": 5.76639503298409, "step": 7430}, {"loss": 1.0837, "grad_norm": 1.2540549039840698, "learning_rate": 0.0002, "epoch": 5.774155995343422, "step": 7440}, {"loss": 1.1527, "grad_norm": 1.5295120477676392, "learning_rate": 0.0002, "epoch": 5.781916957702755, "step": 7450}, {"loss": 1.1146, "grad_norm": 1.1081160306930542, "learning_rate": 0.0002, "epoch": 5.789677920062088, "step": 7460}, {"loss": 1.1304, "grad_norm": 1.4381253719329834, "learning_rate": 0.0002, "epoch": 5.79743888242142, "step": 7470}, {"loss": 1.0684, "grad_norm": 1.3079341650009155, "learning_rate": 0.0002, "epoch": 5.805199844780753, "step": 7480}, {"loss": 1.0544, "grad_norm": 1.1372792720794678, "learning_rate": 0.0002, "epoch": 5.812960807140085, "step": 7490}, {"loss": 1.1622, "grad_norm": 1.3221744298934937, "learning_rate": 0.0002, "epoch": 5.820721769499418, "step": 7500}, {"loss": 1.1515, "grad_norm": 1.3436939716339111, "learning_rate": 0.0002, "epoch": 5.828482731858751, "step": 7510}, {"loss": 1.1154, "grad_norm": 1.3916879892349243, "learning_rate": 0.0002, "epoch": 5.8362436942180835, "step": 7520}, {"loss": 1.0816, "grad_norm": 1.2463704347610474, "learning_rate": 0.0002, "epoch": 5.8440046565774155, "step": 7530}, {"loss": 1.0745, "grad_norm": 1.097051739692688, "learning_rate": 0.0002, "epoch": 5.851765618936748, "step": 7540}, {"loss": 1.1454, "grad_norm": 1.1554739475250244, "learning_rate": 0.0002, "epoch": 5.85952658129608, "step": 7550}, {"loss": 1.0953, "grad_norm": 1.2384694814682007, "learning_rate": 0.0002, "epoch": 5.867287543655413, "step": 7560}, {"loss": 1.1734, "grad_norm": 1.142815351486206, "learning_rate": 0.0002, "epoch": 5.875048506014746, "step": 7570}, {"loss": 1.162, "grad_norm": 1.3637062311172485, "learning_rate": 0.0002, "epoch": 5.882809468374078, "step": 7580}, {"loss": 1.0781, "grad_norm": 1.2449073791503906, "learning_rate": 0.0002, "epoch": 5.890570430733411, "step": 7590}, {"loss": 1.1191, "grad_norm": 1.358058214187622, "learning_rate": 0.0002, "epoch": 5.898331393092743, "step": 7600}, {"loss": 1.0779, "grad_norm": 1.264655351638794, "learning_rate": 0.0002, "epoch": 5.906092355452076, "step": 7610}, {"loss": 1.1538, "grad_norm": 1.3186019659042358, "learning_rate": 0.0002, "epoch": 5.913853317811409, "step": 7620}, {"loss": 1.1076, "grad_norm": 1.4111460447311401, "learning_rate": 0.0002, "epoch": 5.921614280170742, "step": 7630}, {"loss": 1.1765, "grad_norm": 1.1078972816467285, "learning_rate": 0.0002, "epoch": 5.929375242530074, "step": 7640}, {"loss": 1.1305, "grad_norm": 1.2742213010787964, "learning_rate": 0.0002, "epoch": 5.9371362048894065, "step": 7650}, {"loss": 1.144, "grad_norm": 1.3412781953811646, "learning_rate": 0.0002, "epoch": 5.9448971672487385, "step": 7660}, {"loss": 1.1642, "grad_norm": 1.123005986213684, "learning_rate": 0.0002, "epoch": 5.952658129608071, "step": 7670}, {"loss": 1.0732, "grad_norm": 1.2203444242477417, "learning_rate": 0.0002, "epoch": 5.960419091967404, "step": 7680}, {"loss": 1.158, "grad_norm": 1.341011643409729, "learning_rate": 0.0002, "epoch": 5.968180054326736, "step": 7690}, {"loss": 1.1144, "grad_norm": 1.2689454555511475, "learning_rate": 0.0002, "epoch": 5.975941016686069, "step": 7700}, {"loss": 1.2051, "grad_norm": 1.1518112421035767, "learning_rate": 0.0002, "epoch": 5.983701979045401, "step": 7710}, {"loss": 1.1868, "grad_norm": 1.3698320388793945, "learning_rate": 0.0002, "epoch": 5.991462941404734, "step": 7720}, {"loss": 1.0651, "grad_norm": 1.2812788486480713, "learning_rate": 0.0002, "epoch": 5.999223903764067, "step": 7730}, {"eval_loss": 2.252762794494629, "eval_runtime": 114.8471, "eval_samples_per_second": 4.415, "eval_steps_per_second": 0.557, "epoch": 6.0, "step": 7731}, {"loss": 0.8629, "grad_norm": 1.8642009496688843, "learning_rate": 0.0002, "epoch": 6.006984866123399, "step": 7740}, {"loss": 0.8435, "grad_norm": 1.7081232070922852, "learning_rate": 0.0002, "epoch": 6.014745828482732, "step": 7750}, {"loss": 0.7729, "grad_norm": 1.6233899593353271, "learning_rate": 0.0002, "epoch": 6.022506790842065, "step": 7760}, {"loss": 0.7907, "grad_norm": 1.5111888647079468, "learning_rate": 0.0002, "epoch": 6.030267753201397, "step": 7770}, {"loss": 0.7908, "grad_norm": 1.5278418064117432, "learning_rate": 0.0002, "epoch": 6.0380287155607295, "step": 7780}, {"loss": 0.835, "grad_norm": 1.5932185649871826, "learning_rate": 0.0002, "epoch": 6.045789677920062, "step": 7790}, {"loss": 0.7682, "grad_norm": 1.5990597009658813, "learning_rate": 0.0002, "epoch": 6.053550640279394, "step": 7800}, {"loss": 0.8559, "grad_norm": 1.7498669624328613, "learning_rate": 0.0002, "epoch": 6.061311602638727, "step": 7810}, {"loss": 0.8069, "grad_norm": 1.6105555295944214, "learning_rate": 0.0002, "epoch": 6.06907256499806, "step": 7820}, {"loss": 0.8473, "grad_norm": 1.5214293003082275, "learning_rate": 0.0002, "epoch": 6.076833527357392, "step": 7830}, {"loss": 0.8328, "grad_norm": 1.6586973667144775, "learning_rate": 0.0002, "epoch": 6.084594489716725, "step": 7840}, {"loss": 0.8415, "grad_norm": 1.467391848564148, "learning_rate": 0.0002, "epoch": 6.092355452076057, "step": 7850}, {"loss": 0.8274, "grad_norm": 1.537361741065979, "learning_rate": 0.0002, "epoch": 6.10011641443539, "step": 7860}, {"loss": 0.8011, "grad_norm": 1.621764898300171, "learning_rate": 0.0002, "epoch": 6.107877376794723, "step": 7870}, {"loss": 0.8556, "grad_norm": 1.583751916885376, "learning_rate": 0.0002, "epoch": 6.115638339154055, "step": 7880}, {"loss": 0.8829, "grad_norm": 1.6199619770050049, "learning_rate": 0.0002, "epoch": 6.123399301513388, "step": 7890}, {"loss": 0.8226, "grad_norm": 1.6163095235824585, "learning_rate": 0.0002, "epoch": 6.1311602638727205, "step": 7900}, {"loss": 0.8203, "grad_norm": 1.6120976209640503, "learning_rate": 0.0002, "epoch": 6.1389212262320525, "step": 7910}, {"loss": 0.7915, "grad_norm": 1.7886850833892822, "learning_rate": 0.0002, "epoch": 6.146682188591385, "step": 7920}, {"loss": 0.7808, "grad_norm": 1.408303141593933, "learning_rate": 0.0002, "epoch": 6.154443150950718, "step": 7930}, {"loss": 0.8404, "grad_norm": 1.6048113107681274, "learning_rate": 0.0002, "epoch": 6.16220411331005, "step": 7940}, {"loss": 0.8705, "grad_norm": 1.424306869506836, "learning_rate": 0.0002, "epoch": 6.169965075669383, "step": 7950}, {"loss": 0.8177, "grad_norm": 1.4453672170639038, "learning_rate": 0.0002, "epoch": 6.177726038028716, "step": 7960}, {"loss": 0.8182, "grad_norm": 1.3157061338424683, "learning_rate": 0.0002, "epoch": 6.185487000388048, "step": 7970}, {"loss": 0.891, "grad_norm": 1.330541729927063, "learning_rate": 0.0002, "epoch": 6.193247962747381, "step": 7980}, {"loss": 0.8599, "grad_norm": 1.6306229829788208, "learning_rate": 0.0002, "epoch": 6.201008925106713, "step": 7990}, {"loss": 0.9069, "grad_norm": 1.6332136392593384, "learning_rate": 0.0002, "epoch": 6.208769887466046, "step": 8000}, {"loss": 0.83, "grad_norm": 1.708613395690918, "learning_rate": 0.0002, "epoch": 6.216530849825379, "step": 8010}, {"loss": 0.8509, "grad_norm": 1.6637346744537354, "learning_rate": 0.0002, "epoch": 6.224291812184711, "step": 8020}, {"loss": 0.84, "grad_norm": 1.5675315856933594, "learning_rate": 0.0002, "epoch": 6.2320527745440435, "step": 8030}, {"loss": 0.8491, "grad_norm": 1.5826327800750732, "learning_rate": 0.0002, "epoch": 6.239813736903376, "step": 8040}, {"loss": 0.8374, "grad_norm": 1.7382984161376953, "learning_rate": 0.0002, "epoch": 6.247574699262708, "step": 8050}, {"loss": 0.8795, "grad_norm": 1.5272295475006104, "learning_rate": 0.0002, "epoch": 6.255335661622041, "step": 8060}, {"loss": 0.8745, "grad_norm": 1.8195022344589233, "learning_rate": 0.0002, "epoch": 6.263096623981374, "step": 8070}, {"loss": 0.8743, "grad_norm": 1.679901361465454, "learning_rate": 0.0002, "epoch": 6.270857586340706, "step": 8080}, {"loss": 0.9006, "grad_norm": 1.4921348094940186, "learning_rate": 0.0002, "epoch": 6.278618548700039, "step": 8090}, {"loss": 0.899, "grad_norm": 1.4627857208251953, "learning_rate": 0.0002, "epoch": 6.286379511059371, "step": 8100}, {"loss": 0.8944, "grad_norm": 1.3528631925582886, "learning_rate": 0.0002, "epoch": 6.294140473418704, "step": 8110}, {"loss": 0.9355, "grad_norm": 1.6863102912902832, "learning_rate": 0.0002, "epoch": 6.301901435778037, "step": 8120}, {"loss": 0.8764, "grad_norm": 1.6178052425384521, "learning_rate": 0.0002, "epoch": 6.309662398137369, "step": 8130}, {"loss": 0.9182, "grad_norm": 1.7626280784606934, "learning_rate": 0.0002, "epoch": 6.317423360496702, "step": 8140}, {"loss": 0.8886, "grad_norm": 1.7188845872879028, "learning_rate": 0.0002, "epoch": 6.3251843228560345, "step": 8150}, {"loss": 0.895, "grad_norm": 1.5777133703231812, "learning_rate": 0.0002, "epoch": 6.3329452852153665, "step": 8160}, {"loss": 0.9247, "grad_norm": 1.7653207778930664, "learning_rate": 0.0002, "epoch": 6.340706247574699, "step": 8170}, {"loss": 0.8003, "grad_norm": 1.6861237287521362, "learning_rate": 0.0002, "epoch": 6.348467209934032, "step": 8180}, {"loss": 0.884, "grad_norm": 1.6318124532699585, "learning_rate": 0.0002, "epoch": 6.356228172293364, "step": 8190}, {"loss": 0.8341, "grad_norm": 1.6192939281463623, "learning_rate": 0.0002, "epoch": 6.363989134652697, "step": 8200}, {"loss": 0.8939, "grad_norm": 1.7641773223876953, "learning_rate": 0.0002, "epoch": 6.371750097012029, "step": 8210}, {"loss": 0.8582, "grad_norm": 1.6470493078231812, "learning_rate": 0.0002, "epoch": 6.379511059371362, "step": 8220}, {"loss": 0.9351, "grad_norm": 1.5898468494415283, "learning_rate": 0.0002, "epoch": 6.387272021730695, "step": 8230}, {"loss": 0.9658, "grad_norm": 1.8025981187820435, "learning_rate": 0.0002, "epoch": 6.395032984090027, "step": 8240}, {"loss": 0.8953, "grad_norm": 1.7035106420516968, "learning_rate": 0.0002, "epoch": 6.40279394644936, "step": 8250}, {"loss": 0.9193, "grad_norm": 1.5968799591064453, "learning_rate": 0.0002, "epoch": 6.410554908808693, "step": 8260}, {"loss": 0.929, "grad_norm": 1.7492800951004028, "learning_rate": 0.0002, "epoch": 6.418315871168025, "step": 8270}, {"loss": 0.9297, "grad_norm": 1.6914138793945312, "learning_rate": 0.0002, "epoch": 6.4260768335273575, "step": 8280}, {"loss": 0.8878, "grad_norm": 1.5761380195617676, "learning_rate": 0.0002, "epoch": 6.43383779588669, "step": 8290}, {"loss": 0.8761, "grad_norm": 1.5164411067962646, "learning_rate": 0.0002, "epoch": 6.441598758246022, "step": 8300}, {"loss": 0.88, "grad_norm": 1.6600215435028076, "learning_rate": 0.0002, "epoch": 6.449359720605355, "step": 8310}, {"loss": 0.9113, "grad_norm": 1.2477679252624512, "learning_rate": 0.0002, "epoch": 6.457120682964687, "step": 8320}, {"loss": 0.8822, "grad_norm": 1.3698599338531494, "learning_rate": 0.0002, "epoch": 6.46488164532402, "step": 8330}, {"loss": 0.9295, "grad_norm": 1.4847341775894165, "learning_rate": 0.0002, "epoch": 6.472642607683353, "step": 8340}, {"loss": 0.9243, "grad_norm": 1.4713412523269653, "learning_rate": 0.0002, "epoch": 6.480403570042685, "step": 8350}, {"loss": 0.9102, "grad_norm": 1.334523320198059, "learning_rate": 0.0002, "epoch": 6.488164532402018, "step": 8360}, {"loss": 0.8563, "grad_norm": 2.0054359436035156, "learning_rate": 0.0002, "epoch": 6.495925494761351, "step": 8370}, {"loss": 0.9759, "grad_norm": 1.560014247894287, "learning_rate": 0.0002, "epoch": 6.503686457120683, "step": 8380}, {"loss": 0.8542, "grad_norm": 1.518526554107666, "learning_rate": 0.0002, "epoch": 6.511447419480016, "step": 8390}, {"loss": 0.937, "grad_norm": 1.3841272592544556, "learning_rate": 0.0002, "epoch": 6.5192083818393485, "step": 8400}, {"loss": 0.9576, "grad_norm": 1.5191527605056763, "learning_rate": 0.0002, "epoch": 6.5269693441986805, "step": 8410}, {"loss": 0.8899, "grad_norm": 1.5275579690933228, "learning_rate": 0.0002, "epoch": 6.534730306558013, "step": 8420}, {"loss": 0.9291, "grad_norm": 1.621590256690979, "learning_rate": 0.0002, "epoch": 6.542491268917345, "step": 8430}, {"loss": 0.9011, "grad_norm": 1.7939082384109497, "learning_rate": 0.0002, "epoch": 6.550252231276678, "step": 8440}, {"loss": 0.8896, "grad_norm": 1.4542964696884155, "learning_rate": 0.0002, "epoch": 6.558013193636011, "step": 8450}, {"loss": 0.9393, "grad_norm": 1.5458455085754395, "learning_rate": 0.0002, "epoch": 6.565774155995343, "step": 8460}, {"loss": 0.9028, "grad_norm": 1.550359845161438, "learning_rate": 0.0002, "epoch": 6.573535118354676, "step": 8470}, {"loss": 0.9271, "grad_norm": 1.527757167816162, "learning_rate": 0.0002, "epoch": 6.581296080714009, "step": 8480}, {"loss": 0.966, "grad_norm": 1.4683486223220825, "learning_rate": 0.0002, "epoch": 6.589057043073341, "step": 8490}, {"loss": 0.9079, "grad_norm": 1.5057084560394287, "learning_rate": 0.0002, "epoch": 6.596818005432674, "step": 8500}, {"loss": 0.9235, "grad_norm": 1.648289442062378, "learning_rate": 0.0002, "epoch": 6.604578967792007, "step": 8510}, {"loss": 0.9113, "grad_norm": 1.578914761543274, "learning_rate": 0.0002, "epoch": 6.612339930151339, "step": 8520}, {"loss": 0.8894, "grad_norm": 1.5064080953598022, "learning_rate": 0.0002, "epoch": 6.6201008925106715, "step": 8530}, {"loss": 0.8981, "grad_norm": 1.5717744827270508, "learning_rate": 0.0002, "epoch": 6.6278618548700035, "step": 8540}, {"loss": 0.887, "grad_norm": 1.7954767942428589, "learning_rate": 0.0002, "epoch": 6.635622817229336, "step": 8550}, {"loss": 0.927, "grad_norm": 1.6172343492507935, "learning_rate": 0.0002, "epoch": 6.643383779588669, "step": 8560}, {"loss": 0.9384, "grad_norm": 1.6627886295318604, "learning_rate": 0.0002, "epoch": 6.651144741948001, "step": 8570}, {"loss": 0.959, "grad_norm": 1.5264919996261597, "learning_rate": 0.0002, "epoch": 6.658905704307334, "step": 8580}, {"loss": 0.9103, "grad_norm": 1.609248161315918, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 8590}, {"loss": 0.9395, "grad_norm": 1.5474581718444824, "learning_rate": 0.0002, "epoch": 6.674427629025999, "step": 8600}, {"loss": 0.9018, "grad_norm": 1.6294898986816406, "learning_rate": 0.0002, "epoch": 6.682188591385332, "step": 8610}, {"loss": 0.9323, "grad_norm": 1.612615942955017, "learning_rate": 0.0002, "epoch": 6.689949553744665, "step": 8620}, {"loss": 0.9218, "grad_norm": 1.741325855255127, "learning_rate": 0.0002, "epoch": 6.697710516103997, "step": 8630}, {"loss": 1.0475, "grad_norm": 1.5089004039764404, "learning_rate": 0.0002, "epoch": 6.70547147846333, "step": 8640}, {"loss": 1.0009, "grad_norm": 1.4725582599639893, "learning_rate": 0.0002, "epoch": 6.713232440822662, "step": 8650}, {"loss": 0.9818, "grad_norm": 1.6992095708847046, "learning_rate": 0.0002, "epoch": 6.7209934031819945, "step": 8660}, {"loss": 0.9229, "grad_norm": 1.5938470363616943, "learning_rate": 0.0002, "epoch": 6.728754365541327, "step": 8670}, {"loss": 0.9411, "grad_norm": 1.58723783493042, "learning_rate": 0.0002, "epoch": 6.736515327900659, "step": 8680}, {"loss": 0.9738, "grad_norm": 1.514389991760254, "learning_rate": 0.0002, "epoch": 6.744276290259992, "step": 8690}, {"loss": 0.9283, "grad_norm": 1.6799157857894897, "learning_rate": 0.0002, "epoch": 6.752037252619325, "step": 8700}, {"loss": 0.9138, "grad_norm": 1.5436359643936157, "learning_rate": 0.0002, "epoch": 6.759798214978657, "step": 8710}, {"loss": 0.9678, "grad_norm": 1.477137565612793, "learning_rate": 0.0002, "epoch": 6.76755917733799, "step": 8720}, {"loss": 1.0044, "grad_norm": 1.7383503913879395, "learning_rate": 0.0002, "epoch": 6.775320139697323, "step": 8730}, {"loss": 0.9492, "grad_norm": 1.8000324964523315, "learning_rate": 0.0002, "epoch": 6.783081102056655, "step": 8740}, {"loss": 0.8943, "grad_norm": 1.3099453449249268, "learning_rate": 0.0002, "epoch": 6.790842064415988, "step": 8750}, {"loss": 0.9709, "grad_norm": 1.8775172233581543, "learning_rate": 0.0002, "epoch": 6.79860302677532, "step": 8760}, {"loss": 0.9356, "grad_norm": 1.5832085609436035, "learning_rate": 0.0002, "epoch": 6.806363989134653, "step": 8770}, {"loss": 0.9397, "grad_norm": 1.4903252124786377, "learning_rate": 0.0002, "epoch": 6.8141249514939854, "step": 8780}, {"loss": 0.9602, "grad_norm": 1.6360470056533813, "learning_rate": 0.0002, "epoch": 6.821885913853317, "step": 8790}, {"loss": 0.957, "grad_norm": 1.5457707643508911, "learning_rate": 0.0002, "epoch": 6.82964687621265, "step": 8800}, {"loss": 0.943, "grad_norm": 1.5449066162109375, "learning_rate": 0.0002, "epoch": 6.837407838571983, "step": 8810}, {"loss": 1.0007, "grad_norm": 1.618337631225586, "learning_rate": 0.0002, "epoch": 6.845168800931315, "step": 8820}, {"loss": 0.9314, "grad_norm": 1.38296639919281, "learning_rate": 0.0002, "epoch": 6.852929763290648, "step": 8830}, {"loss": 0.9349, "grad_norm": 1.6427991390228271, "learning_rate": 0.0002, "epoch": 6.860690725649981, "step": 8840}, {"loss": 1.0194, "grad_norm": 1.4980270862579346, "learning_rate": 0.0002, "epoch": 6.868451688009313, "step": 8850}, {"loss": 0.9541, "grad_norm": 1.3800020217895508, "learning_rate": 0.0002, "epoch": 6.876212650368646, "step": 8860}, {"loss": 1.0102, "grad_norm": 1.5971838235855103, "learning_rate": 0.0002, "epoch": 6.883973612727978, "step": 8870}, {"loss": 1.0105, "grad_norm": 1.4429489374160767, "learning_rate": 0.0002, "epoch": 6.891734575087311, "step": 8880}, {"loss": 0.9143, "grad_norm": 1.4959166049957275, "learning_rate": 0.0002, "epoch": 6.899495537446644, "step": 8890}, {"loss": 0.9403, "grad_norm": 1.5776222944259644, "learning_rate": 0.0002, "epoch": 6.907256499805976, "step": 8900}, {"loss": 0.9256, "grad_norm": 1.510412573814392, "learning_rate": 0.0002, "epoch": 6.915017462165308, "step": 8910}, {"loss": 1.0095, "grad_norm": 1.7216295003890991, "learning_rate": 0.0002, "epoch": 6.922778424524641, "step": 8920}, {"loss": 0.9464, "grad_norm": 1.830762505531311, "learning_rate": 0.0002, "epoch": 6.930539386883973, "step": 8930}, {"loss": 0.9704, "grad_norm": 1.3472434282302856, "learning_rate": 0.0002, "epoch": 6.938300349243306, "step": 8940}, {"loss": 0.9718, "grad_norm": 1.5748040676116943, "learning_rate": 0.0002, "epoch": 6.946061311602639, "step": 8950}, {"loss": 0.9891, "grad_norm": 1.5317506790161133, "learning_rate": 0.0002, "epoch": 6.953822273961971, "step": 8960}, {"loss": 0.9513, "grad_norm": 1.5565721988677979, "learning_rate": 0.0002, "epoch": 6.961583236321304, "step": 8970}, {"loss": 0.9118, "grad_norm": 1.5288970470428467, "learning_rate": 0.0002, "epoch": 6.969344198680636, "step": 8980}, {"loss": 0.9789, "grad_norm": 1.562624454498291, "learning_rate": 0.0002, "epoch": 6.977105161039969, "step": 8990}, {"loss": 0.9929, "grad_norm": 1.3777633905410767, "learning_rate": 0.0002, "epoch": 6.984866123399302, "step": 9000}, {"loss": 0.9713, "grad_norm": 1.5868972539901733, "learning_rate": 0.0002, "epoch": 6.992627085758635, "step": 9010}, {"eval_loss": 2.4372169971466064, "eval_runtime": 113.8966, "eval_samples_per_second": 4.451, "eval_steps_per_second": 0.562, "epoch": 6.9996119518820334, "step": 9019}, {"loss": 0.9911, "grad_norm": 1.3035310506820679, "learning_rate": 0.0002, "epoch": 7.0003880481179666, "step": 9020}, {"loss": 0.6585, "grad_norm": 1.5640218257904053, "learning_rate": 0.0002, "epoch": 7.008149010477299, "step": 9030}, {"loss": 0.6507, "grad_norm": 1.9529098272323608, "learning_rate": 0.0002, "epoch": 7.015909972836631, "step": 9040}, {"loss": 0.6335, "grad_norm": 1.6257457733154297, "learning_rate": 0.0002, "epoch": 7.023670935195964, "step": 9050}, {"loss": 0.6752, "grad_norm": 1.8028602600097656, "learning_rate": 0.0002, "epoch": 7.031431897555297, "step": 9060}, {"loss": 0.66, "grad_norm": 1.4882043600082397, "learning_rate": 0.0002, "epoch": 7.039192859914629, "step": 9070}, {"loss": 0.7117, "grad_norm": 2.10062837600708, "learning_rate": 0.0002, "epoch": 7.046953822273962, "step": 9080}, {"loss": 0.643, "grad_norm": 1.6754050254821777, "learning_rate": 0.0002, "epoch": 7.054714784633295, "step": 9090}, {"loss": 0.6461, "grad_norm": 1.9425220489501953, "learning_rate": 0.0002, "epoch": 7.062475746992627, "step": 9100}, {"loss": 0.6441, "grad_norm": 1.9451842308044434, "learning_rate": 0.0002, "epoch": 7.07023670935196, "step": 9110}, {"loss": 0.6872, "grad_norm": 2.203806161880493, "learning_rate": 0.0002, "epoch": 7.077997671711292, "step": 9120}, {"loss": 0.6768, "grad_norm": 1.7248806953430176, "learning_rate": 0.0002, "epoch": 7.085758634070625, "step": 9130}, {"loss": 0.6911, "grad_norm": 1.5713436603546143, "learning_rate": 0.0002, "epoch": 7.0935195964299576, "step": 9140}, {"loss": 0.6689, "grad_norm": 2.0308637619018555, "learning_rate": 0.0002, "epoch": 7.1012805587892895, "step": 9150}, {"loss": 0.6617, "grad_norm": 1.7522761821746826, "learning_rate": 0.0002, "epoch": 7.109041521148622, "step": 9160}, {"loss": 0.669, "grad_norm": 1.9685192108154297, "learning_rate": 0.0002, "epoch": 7.116802483507955, "step": 9170}, {"loss": 0.6727, "grad_norm": 2.069120407104492, "learning_rate": 0.0002, "epoch": 7.124563445867287, "step": 9180}, {"loss": 0.6553, "grad_norm": 1.7211129665374756, "learning_rate": 0.0002, "epoch": 7.13232440822662, "step": 9190}, {"loss": 0.6827, "grad_norm": 1.7535923719406128, "learning_rate": 0.0002, "epoch": 7.140085370585953, "step": 9200}, {"loss": 0.6698, "grad_norm": 1.7181583642959595, "learning_rate": 0.0002, "epoch": 7.147846332945285, "step": 9210}, {"loss": 0.7056, "grad_norm": 1.7778624296188354, "learning_rate": 0.0002, "epoch": 7.155607295304618, "step": 9220}, {"loss": 0.7326, "grad_norm": 1.8051576614379883, "learning_rate": 0.0002, "epoch": 7.16336825766395, "step": 9230}, {"loss": 0.6858, "grad_norm": 1.9704501628875732, "learning_rate": 0.0002, "epoch": 7.171129220023283, "step": 9240}, {"loss": 0.7029, "grad_norm": 1.8750483989715576, "learning_rate": 0.0002, "epoch": 7.178890182382616, "step": 9250}, {"loss": 0.7073, "grad_norm": 1.790107011795044, "learning_rate": 0.0002, "epoch": 7.186651144741948, "step": 9260}, {"loss": 0.6875, "grad_norm": 1.9681477546691895, "learning_rate": 0.0002, "epoch": 7.1944121071012805, "step": 9270}, {"loss": 0.7074, "grad_norm": 1.7811331748962402, "learning_rate": 0.0002, "epoch": 7.202173069460613, "step": 9280}, {"loss": 0.6959, "grad_norm": 1.9861894845962524, "learning_rate": 0.0002, "epoch": 7.209934031819945, "step": 9290}, {"loss": 0.6463, "grad_norm": 1.8615444898605347, "learning_rate": 0.0002, "epoch": 7.217694994179278, "step": 9300}, {"loss": 0.7071, "grad_norm": 1.899372935295105, "learning_rate": 0.0002, "epoch": 7.225455956538611, "step": 9310}, {"loss": 0.6883, "grad_norm": 1.8131160736083984, "learning_rate": 0.0002, "epoch": 7.233216918897943, "step": 9320}, {"loss": 0.6813, "grad_norm": 1.5020049810409546, "learning_rate": 0.0002, "epoch": 7.240977881257276, "step": 9330}, {"loss": 0.7197, "grad_norm": 1.7210577726364136, "learning_rate": 0.0002, "epoch": 7.248738843616608, "step": 9340}, {"loss": 0.6769, "grad_norm": 2.003021001815796, "learning_rate": 0.0002, "epoch": 7.256499805975941, "step": 9350}, {"loss": 0.6746, "grad_norm": 1.6632959842681885, "learning_rate": 0.0002, "epoch": 7.264260768335274, "step": 9360}, {"loss": 0.7072, "grad_norm": 1.7717185020446777, "learning_rate": 0.0002, "epoch": 7.272021730694606, "step": 9370}, {"loss": 0.7447, "grad_norm": 1.8554900884628296, "learning_rate": 0.0002, "epoch": 7.279782693053939, "step": 9380}, {"loss": 0.6983, "grad_norm": 1.889708399772644, "learning_rate": 0.0002, "epoch": 7.2875436554132715, "step": 9390}, {"loss": 0.7495, "grad_norm": 1.8426263332366943, "learning_rate": 0.0002, "epoch": 7.2953046177726035, "step": 9400}, {"loss": 0.6854, "grad_norm": 1.9665130376815796, "learning_rate": 0.0002, "epoch": 7.303065580131936, "step": 9410}, {"loss": 0.6466, "grad_norm": 1.9337282180786133, "learning_rate": 0.0002, "epoch": 7.310826542491269, "step": 9420}, {"loss": 0.6885, "grad_norm": 1.8582539558410645, "learning_rate": 0.0002, "epoch": 7.318587504850601, "step": 9430}, {"loss": 0.7366, "grad_norm": 1.8865947723388672, "learning_rate": 0.0002, "epoch": 7.326348467209934, "step": 9440}, {"loss": 0.7825, "grad_norm": 1.8144744634628296, "learning_rate": 0.0002, "epoch": 7.334109429569267, "step": 9450}, {"loss": 0.6706, "grad_norm": 1.6930001974105835, "learning_rate": 0.0002, "epoch": 7.341870391928599, "step": 9460}, {"loss": 0.7502, "grad_norm": 1.7389107942581177, "learning_rate": 0.0002, "epoch": 7.349631354287932, "step": 9470}, {"loss": 0.7264, "grad_norm": 1.6860785484313965, "learning_rate": 0.0002, "epoch": 7.357392316647264, "step": 9480}, {"loss": 0.7593, "grad_norm": 2.2142202854156494, "learning_rate": 0.0002, "epoch": 7.365153279006597, "step": 9490}, {"loss": 0.7936, "grad_norm": 1.9988185167312622, "learning_rate": 0.0002, "epoch": 7.37291424136593, "step": 9500}, {"loss": 0.7348, "grad_norm": 1.7517266273498535, "learning_rate": 0.0002, "epoch": 7.380675203725262, "step": 9510}, {"loss": 0.7432, "grad_norm": 1.7426788806915283, "learning_rate": 0.0002, "epoch": 7.3884361660845945, "step": 9520}, {"loss": 0.7649, "grad_norm": 1.8157157897949219, "learning_rate": 0.0002, "epoch": 7.396197128443927, "step": 9530}, {"loss": 0.716, "grad_norm": 1.9826514720916748, "learning_rate": 0.0002, "epoch": 7.403958090803259, "step": 9540}, {"loss": 0.7685, "grad_norm": 1.9057488441467285, "learning_rate": 0.0002, "epoch": 7.411719053162592, "step": 9550}, {"loss": 0.7709, "grad_norm": 2.0701088905334473, "learning_rate": 0.0002, "epoch": 7.419480015521925, "step": 9560}, {"loss": 0.7438, "grad_norm": 1.8623783588409424, "learning_rate": 0.0002, "epoch": 7.427240977881257, "step": 9570}, {"loss": 0.7719, "grad_norm": 1.780672550201416, "learning_rate": 0.0002, "epoch": 7.43500194024059, "step": 9580}, {"loss": 0.7579, "grad_norm": 1.8437316417694092, "learning_rate": 0.0002, "epoch": 7.442762902599922, "step": 9590}, {"loss": 0.8044, "grad_norm": 1.9327329397201538, "learning_rate": 0.0002, "epoch": 7.450523864959255, "step": 9600}, {"loss": 0.7212, "grad_norm": 2.011535167694092, "learning_rate": 0.0002, "epoch": 7.458284827318588, "step": 9610}, {"loss": 0.7586, "grad_norm": 1.7557756900787354, "learning_rate": 0.0002, "epoch": 7.46604578967792, "step": 9620}, {"loss": 0.7461, "grad_norm": 1.8508951663970947, "learning_rate": 0.0002, "epoch": 7.473806752037253, "step": 9630}, {"loss": 0.7425, "grad_norm": 1.8236663341522217, "learning_rate": 0.0002, "epoch": 7.4815677143965855, "step": 9640}, {"loss": 0.7107, "grad_norm": 1.9308991432189941, "learning_rate": 0.0002, "epoch": 7.4893286767559175, "step": 9650}, {"loss": 0.738, "grad_norm": 1.8095420598983765, "learning_rate": 0.0002, "epoch": 7.49708963911525, "step": 9660}, {"loss": 0.7548, "grad_norm": 1.8216804265975952, "learning_rate": 0.0002, "epoch": 7.504850601474583, "step": 9670}, {"loss": 0.7494, "grad_norm": 1.8275913000106812, "learning_rate": 0.0002, "epoch": 7.512611563833915, "step": 9680}, {"loss": 0.7178, "grad_norm": 1.8217864036560059, "learning_rate": 0.0002, "epoch": 7.520372526193248, "step": 9690}, {"loss": 0.7331, "grad_norm": 1.461728572845459, "learning_rate": 0.0002, "epoch": 7.52813348855258, "step": 9700}, {"loss": 0.7597, "grad_norm": 1.6569337844848633, "learning_rate": 0.0002, "epoch": 7.535894450911913, "step": 9710}, {"loss": 0.8117, "grad_norm": 2.332101821899414, "learning_rate": 0.0002, "epoch": 7.543655413271246, "step": 9720}, {"loss": 0.7937, "grad_norm": 1.760307788848877, "learning_rate": 0.0002, "epoch": 7.551416375630578, "step": 9730}, {"loss": 0.7213, "grad_norm": 2.2455198764801025, "learning_rate": 0.0002, "epoch": 7.559177337989911, "step": 9740}, {"loss": 0.8122, "grad_norm": 1.819676399230957, "learning_rate": 0.0002, "epoch": 7.566938300349244, "step": 9750}, {"loss": 0.754, "grad_norm": 1.579603910446167, "learning_rate": 0.0002, "epoch": 7.574699262708576, "step": 9760}, {"loss": 0.7608, "grad_norm": 1.9687446355819702, "learning_rate": 0.0002, "epoch": 7.5824602250679085, "step": 9770}, {"loss": 0.813, "grad_norm": 1.859029769897461, "learning_rate": 0.0002, "epoch": 7.590221187427241, "step": 9780}, {"loss": 0.7353, "grad_norm": 1.652137279510498, "learning_rate": 0.0002, "epoch": 7.597982149786573, "step": 9790}, {"loss": 0.7388, "grad_norm": 1.6381458044052124, "learning_rate": 0.0002, "epoch": 7.605743112145906, "step": 9800}, {"loss": 0.8265, "grad_norm": 1.8706856966018677, "learning_rate": 0.0002, "epoch": 7.613504074505238, "step": 9810}, {"loss": 0.7223, "grad_norm": 1.8709853887557983, "learning_rate": 0.0002, "epoch": 7.621265036864571, "step": 9820}, {"loss": 0.7972, "grad_norm": 2.1183695793151855, "learning_rate": 0.0002, "epoch": 7.629025999223904, "step": 9830}, {"loss": 0.7645, "grad_norm": 1.6770579814910889, "learning_rate": 0.0002, "epoch": 7.636786961583236, "step": 9840}, {"loss": 0.8197, "grad_norm": 2.0952677726745605, "learning_rate": 0.0002, "epoch": 7.644547923942569, "step": 9850}, {"loss": 0.8483, "grad_norm": 1.7378171682357788, "learning_rate": 0.0002, "epoch": 7.652308886301902, "step": 9860}, {"loss": 0.7942, "grad_norm": 2.0860157012939453, "learning_rate": 0.0002, "epoch": 7.660069848661234, "step": 9870}, {"loss": 0.769, "grad_norm": 1.6986967325210571, "learning_rate": 0.0002, "epoch": 7.667830811020567, "step": 9880}, {"loss": 0.7626, "grad_norm": 1.666702151298523, "learning_rate": 0.0002, "epoch": 7.6755917733798995, "step": 9890}, {"loss": 0.7418, "grad_norm": 2.446931838989258, "learning_rate": 0.0002, "epoch": 7.6833527357392315, "step": 9900}, {"loss": 0.8333, "grad_norm": 1.6984577178955078, "learning_rate": 0.0002, "epoch": 7.691113698098564, "step": 9910}, {"loss": 0.7822, "grad_norm": 2.0732316970825195, "learning_rate": 0.0002, "epoch": 7.698874660457896, "step": 9920}, {"loss": 0.7342, "grad_norm": 1.9884073734283447, "learning_rate": 0.0002, "epoch": 7.706635622817229, "step": 9930}, {"loss": 0.7825, "grad_norm": 1.8669427633285522, "learning_rate": 0.0002, "epoch": 7.714396585176562, "step": 9940}, {"loss": 0.7722, "grad_norm": 1.6163996458053589, "learning_rate": 0.0002, "epoch": 7.722157547535894, "step": 9950}, {"loss": 0.8069, "grad_norm": 2.4492127895355225, "learning_rate": 0.0002, "epoch": 7.729918509895227, "step": 9960}, {"loss": 0.7924, "grad_norm": 1.5625537633895874, "learning_rate": 0.0002, "epoch": 7.73767947225456, "step": 9970}, {"loss": 0.8273, "grad_norm": 1.8205251693725586, "learning_rate": 0.0002, "epoch": 7.745440434613892, "step": 9980}, {"loss": 0.7892, "grad_norm": 1.9902361631393433, "learning_rate": 0.0002, "epoch": 7.753201396973225, "step": 9990}, {"loss": 0.7871, "grad_norm": 1.8294041156768799, "learning_rate": 0.0002, "epoch": 7.760962359332558, "step": 10000}, {"loss": 0.771, "grad_norm": 1.8938281536102295, "learning_rate": 0.0002, "epoch": 7.76872332169189, "step": 10010}, {"loss": 0.8304, "grad_norm": 1.9367564916610718, "learning_rate": 0.0002, "epoch": 7.7764842840512225, "step": 10020}, {"loss": 0.7437, "grad_norm": 1.9130750894546509, "learning_rate": 0.0002, "epoch": 7.7842452464105545, "step": 10030}, {"loss": 0.8359, "grad_norm": 1.931505799293518, "learning_rate": 0.0002, "epoch": 7.792006208769887, "step": 10040}, {"loss": 0.7708, "grad_norm": 2.107954740524292, "learning_rate": 0.0002, "epoch": 7.79976717112922, "step": 10050}, {"loss": 0.7922, "grad_norm": 1.8736399412155151, "learning_rate": 0.0002, "epoch": 7.807528133488552, "step": 10060}, {"loss": 0.8143, "grad_norm": 1.6943566799163818, "learning_rate": 0.0002, "epoch": 7.815289095847885, "step": 10070}, {"loss": 0.7645, "grad_norm": 2.365346908569336, "learning_rate": 0.0002, "epoch": 7.823050058207218, "step": 10080}, {"loss": 0.815, "grad_norm": 2.0044665336608887, "learning_rate": 0.0002, "epoch": 7.83081102056655, "step": 10090}, {"loss": 0.8297, "grad_norm": 1.7680124044418335, "learning_rate": 0.0002, "epoch": 7.838571982925883, "step": 10100}, {"loss": 0.7729, "grad_norm": 1.8494547605514526, "learning_rate": 0.0002, "epoch": 7.846332945285216, "step": 10110}, {"loss": 0.7909, "grad_norm": 1.822107195854187, "learning_rate": 0.0002, "epoch": 7.854093907644548, "step": 10120}, {"loss": 0.8149, "grad_norm": 1.8191527128219604, "learning_rate": 0.0002, "epoch": 7.861854870003881, "step": 10130}, {"loss": 0.8045, "grad_norm": 1.8011466264724731, "learning_rate": 0.0002, "epoch": 7.869615832363213, "step": 10140}, {"loss": 0.79, "grad_norm": 1.7087100744247437, "learning_rate": 0.0002, "epoch": 7.8773767947225455, "step": 10150}, {"loss": 0.8081, "grad_norm": 1.7698721885681152, "learning_rate": 0.0002, "epoch": 7.885137757081878, "step": 10160}, {"loss": 0.7365, "grad_norm": 1.9578531980514526, "learning_rate": 0.0002, "epoch": 7.89289871944121, "step": 10170}, {"loss": 0.8497, "grad_norm": 1.7660179138183594, "learning_rate": 0.0002, "epoch": 7.900659681800543, "step": 10180}, {"loss": 0.7858, "grad_norm": 2.0122673511505127, "learning_rate": 0.0002, "epoch": 7.908420644159876, "step": 10190}, {"loss": 0.7607, "grad_norm": 1.737443208694458, "learning_rate": 0.0002, "epoch": 7.916181606519208, "step": 10200}, {"loss": 0.8365, "grad_norm": 1.6381052732467651, "learning_rate": 0.0002, "epoch": 7.923942568878541, "step": 10210}, {"loss": 0.7855, "grad_norm": 1.8845038414001465, "learning_rate": 0.0002, "epoch": 7.931703531237874, "step": 10220}, {"loss": 0.8354, "grad_norm": 1.952194333076477, "learning_rate": 0.0002, "epoch": 7.939464493597206, "step": 10230}, {"loss": 0.8428, "grad_norm": 1.7254410982131958, "learning_rate": 0.0002, "epoch": 7.947225455956539, "step": 10240}, {"loss": 0.834, "grad_norm": 2.14776873588562, "learning_rate": 0.0002, "epoch": 7.954986418315871, "step": 10250}, {"loss": 0.8144, "grad_norm": 1.7655725479125977, "learning_rate": 0.0002, "epoch": 7.962747380675204, "step": 10260}, {"loss": 0.8176, "grad_norm": 1.7337331771850586, "learning_rate": 0.0002, "epoch": 7.9705083430345365, "step": 10270}, {"loss": 0.8652, "grad_norm": 1.7742228507995605, "learning_rate": 0.0002, "epoch": 7.9782693053938685, "step": 10280}, {"loss": 0.7788, "grad_norm": 1.892137050628662, "learning_rate": 0.0002, "epoch": 7.986030267753201, "step": 10290}, {"loss": 0.8243, "grad_norm": 1.8636300563812256, "learning_rate": 0.0002, "epoch": 7.993791230112534, "step": 10300}]}