diff --git a/.gitattributes b/.gitattributes
index d8c223681de3764884ebd0a74709e87ba2f3d268..5a06715899cde74d872fb1619303b200ad376343 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -2323,3 +2323,12 @@ Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.2-num-1405-sd-10000/checkpoint-754/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.2-num-1405-sd-10000/checkpoint-942/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.2-num-1405-sd-10000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f72e00008c1f013813170eda945d1f77849e55ea
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b265b08aa3b965c22c1355e68e889f9f7a85e11299bd18f4a4c2e48518f9afa
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f72e00008c1f013813170eda945d1f77849e55ea
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b265b08aa3b965c22c1355e68e889f9f7a85e11299bd18f4a4c2e48518f9afa
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6b500f6a9de2d0bc23e173cdd998fba6ffc4e865
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5f458cd1826277fa3587f31098e7044eba7d2c9574a8cee1d883c31bff7ba2f
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e5d2d9125a7fc6c87e8aa0a6f1bfac23f83c990d
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d569bab937ecbb882df0eb4ff018cb21cd7eafe1de6088e0f040e8630a49579e
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5426131138c9c38d28263f0205d4fc1b8f86900a
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3046e60eecb5c9ef3ecd6e32c300fbab8fd731d5983e3b78f98c89d3dd3c37fb
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..e32b5d85113135fefe642aa1486c0264fd161668
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/trainer_state.json
@@ -0,0 +1,981 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 3.0,
+  "eval_steps": 10,
+  "global_step": 1323,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 1.6169701151342592e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7c71c50bbfa54b065a9947740f68bda108d194d3
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d40dc0076f9408cf439ef597eee90d25c8010d4c697cbdb72719852ccc93301
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b858c112b0b22fe707bde7288282f8f7fc75de0a
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f386c7ebef35dc79f063801ee7f6df708cd069f8df5cca7ad7f88b066b8e4643
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..68490f8d991067d74451d00ec498a1d45eb32d25
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d862ecc16e3b212ed748acf953a6d6fd8467788b13353a8a7f520862115f016
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..86aac02023a02ac1cd4cc108db7ef57ab5a6f571
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c56e47336a16ef9e258bc2b03be58373a485c239c4356f8555fd6d380b5e131c
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3fd7307ba30f446c76c4dc4ab9fd8d99582899b5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/trainer_state.json
@@ -0,0 +1,1297 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 1764,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    },
+    {
+      "epoch": 3.015873015873016,
+      "grad_norm": 0.625606119632721,
+      "learning_rate": 0.0002,
+      "loss": 1.8798,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0385487528344672,
+      "grad_norm": 0.6679854989051819,
+      "learning_rate": 0.0002,
+      "loss": 1.8229,
+      "step": 1340
+    },
+    {
+      "epoch": 3.061224489795918,
+      "grad_norm": 0.6756007671356201,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1350
+    },
+    {
+      "epoch": 3.0839002267573696,
+      "grad_norm": 0.8019710779190063,
+      "learning_rate": 0.0002,
+      "loss": 1.8126,
+      "step": 1360
+    },
+    {
+      "epoch": 3.106575963718821,
+      "grad_norm": 0.9087499976158142,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1370
+    },
+    {
+      "epoch": 3.129251700680272,
+      "grad_norm": 0.7627860903739929,
+      "learning_rate": 0.0002,
+      "loss": 1.8951,
+      "step": 1380
+    },
+    {
+      "epoch": 3.1519274376417235,
+      "grad_norm": 0.761993944644928,
+      "learning_rate": 0.0002,
+      "loss": 1.8308,
+      "step": 1390
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6993107795715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9513,
+      "step": 1400
+    },
+    {
+      "epoch": 3.197278911564626,
+      "grad_norm": 0.6398035287857056,
+      "learning_rate": 0.0002,
+      "loss": 1.8857,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2199546485260773,
+      "grad_norm": 0.8602580428123474,
+      "learning_rate": 0.0002,
+      "loss": 1.9041,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2426303854875282,
+      "grad_norm": 0.7799932360649109,
+      "learning_rate": 0.0002,
+      "loss": 1.9629,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2653061224489797,
+      "grad_norm": 0.6598666906356812,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 1440
+    },
+    {
+      "epoch": 3.287981859410431,
+      "grad_norm": 0.6513776779174805,
+      "learning_rate": 0.0002,
+      "loss": 1.8772,
+      "step": 1450
+    },
+    {
+      "epoch": 3.310657596371882,
+      "grad_norm": 0.7690088748931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8568,
+      "step": 1460
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 1.0070042610168457,
+      "learning_rate": 0.0002,
+      "loss": 1.8535,
+      "step": 1470
+    },
+    {
+      "epoch": 3.3560090702947845,
+      "grad_norm": 0.8606905341148376,
+      "learning_rate": 0.0002,
+      "loss": 1.8971,
+      "step": 1480
+    },
+    {
+      "epoch": 3.378684807256236,
+      "grad_norm": 0.7596806287765503,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4013605442176873,
+      "grad_norm": 0.6961908936500549,
+      "learning_rate": 0.0002,
+      "loss": 1.9278,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4240362811791383,
+      "grad_norm": 0.8368703126907349,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 1510
+    },
+    {
+      "epoch": 3.4467120181405897,
+      "grad_norm": 0.6655045747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.924,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4693877551020407,
+      "grad_norm": 0.7557174563407898,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1530
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.6713934540748596,
+      "learning_rate": 0.0002,
+      "loss": 1.8435,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5147392290249435,
+      "grad_norm": 0.7006916999816895,
+      "learning_rate": 0.0002,
+      "loss": 1.9016,
+      "step": 1550
+    },
+    {
+      "epoch": 3.5374149659863945,
+      "grad_norm": 0.6728846430778503,
+      "learning_rate": 0.0002,
+      "loss": 1.9208,
+      "step": 1560
+    },
+    {
+      "epoch": 3.560090702947846,
+      "grad_norm": 0.7479315400123596,
+      "learning_rate": 0.0002,
+      "loss": 1.8769,
+      "step": 1570
+    },
+    {
+      "epoch": 3.582766439909297,
+      "grad_norm": 0.7596876621246338,
+      "learning_rate": 0.0002,
+      "loss": 1.8558,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6054421768707483,
+      "grad_norm": 0.8948362469673157,
+      "learning_rate": 0.0002,
+      "loss": 1.876,
+      "step": 1590
+    },
+    {
+      "epoch": 3.6281179138321997,
+      "grad_norm": 0.6952481269836426,
+      "learning_rate": 0.0002,
+      "loss": 1.8552,
+      "step": 1600
+    },
+    {
+      "epoch": 3.6507936507936507,
+      "grad_norm": 0.6494414210319519,
+      "learning_rate": 0.0002,
+      "loss": 1.933,
+      "step": 1610
+    },
+    {
+      "epoch": 3.673469387755102,
+      "grad_norm": 0.7438889145851135,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1620
+    },
+    {
+      "epoch": 3.696145124716553,
+      "grad_norm": 0.7470757961273193,
+      "learning_rate": 0.0002,
+      "loss": 1.9441,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7188208616780045,
+      "grad_norm": 0.8727335333824158,
+      "learning_rate": 0.0002,
+      "loss": 1.8685,
+      "step": 1640
+    },
+    {
+      "epoch": 3.741496598639456,
+      "grad_norm": 0.7064452767372131,
+      "learning_rate": 0.0002,
+      "loss": 1.8831,
+      "step": 1650
+    },
+    {
+      "epoch": 3.764172335600907,
+      "grad_norm": 0.6958122849464417,
+      "learning_rate": 0.0002,
+      "loss": 1.9743,
+      "step": 1660
+    },
+    {
+      "epoch": 3.7868480725623583,
+      "grad_norm": 0.6751590371131897,
+      "learning_rate": 0.0002,
+      "loss": 1.927,
+      "step": 1670
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.7274497151374817,
+      "learning_rate": 0.0002,
+      "loss": 1.8868,
+      "step": 1680
+    },
+    {
+      "epoch": 3.8321995464852607,
+      "grad_norm": 0.7311708331108093,
+      "learning_rate": 0.0002,
+      "loss": 1.942,
+      "step": 1690
+    },
+    {
+      "epoch": 3.854875283446712,
+      "grad_norm": 0.6703173518180847,
+      "learning_rate": 0.0002,
+      "loss": 1.8134,
+      "step": 1700
+    },
+    {
+      "epoch": 3.877551020408163,
+      "grad_norm": 0.6263390779495239,
+      "learning_rate": 0.0002,
+      "loss": 1.8687,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9002267573696145,
+      "grad_norm": 0.744022011756897,
+      "learning_rate": 0.0002,
+      "loss": 1.9206,
+      "step": 1720
+    },
+    {
+      "epoch": 3.9229024943310655,
+      "grad_norm": 0.7496722340583801,
+      "learning_rate": 0.0002,
+      "loss": 1.8961,
+      "step": 1730
+    },
+    {
+      "epoch": 3.945578231292517,
+      "grad_norm": 0.6877635717391968,
+      "learning_rate": 0.0002,
+      "loss": 1.9212,
+      "step": 1740
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.711262047290802,
+      "learning_rate": 0.0002,
+      "loss": 1.8504,
+      "step": 1750
+    },
+    {
+      "epoch": 3.9909297052154193,
+      "grad_norm": 0.7452033758163452,
+      "learning_rate": 0.0002,
+      "loss": 1.9315,
+      "step": 1760
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.065720319747925,
+      "eval_runtime": 14.7103,
+      "eval_samples_per_second": 34.466,
+      "eval_steps_per_second": 4.351,
+      "step": 1764
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 2.1559601535123456e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1764/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a580f1e4183d04ca9ada10f3b29c3209519e30e3
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bfa581613822d5b479a2aad8252e23ff88d88624469f545cd35060a48323076
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7b23edf3633a44cbfb4fbefb585516418cbd1b36
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2c38485d449470b16ebc509248b9bf232bbb8a88fea9517a6fdc2b483160e2a
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..905851c11515b970b75d90254d866c1bc7022b4e
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc704d2bc5a01caab7b4a7b5ca1c1a11b4dc79579e11338ae1a603a99cbe42c4
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2372deb995fd297c304157286266a69428ffcf6f
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54c9da8a02a9846b4fa97e500528bf4b60b78e598eeffb8c8538e5ef0a14a2b6
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3539ff4b53f98847e95fdf46d948e62459f8f126
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/trainer_state.json
@@ -0,0 +1,1613 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 5.0,
+  "eval_steps": 10,
+  "global_step": 2205,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    },
+    {
+      "epoch": 3.015873015873016,
+      "grad_norm": 0.625606119632721,
+      "learning_rate": 0.0002,
+      "loss": 1.8798,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0385487528344672,
+      "grad_norm": 0.6679854989051819,
+      "learning_rate": 0.0002,
+      "loss": 1.8229,
+      "step": 1340
+    },
+    {
+      "epoch": 3.061224489795918,
+      "grad_norm": 0.6756007671356201,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1350
+    },
+    {
+      "epoch": 3.0839002267573696,
+      "grad_norm": 0.8019710779190063,
+      "learning_rate": 0.0002,
+      "loss": 1.8126,
+      "step": 1360
+    },
+    {
+      "epoch": 3.106575963718821,
+      "grad_norm": 0.9087499976158142,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1370
+    },
+    {
+      "epoch": 3.129251700680272,
+      "grad_norm": 0.7627860903739929,
+      "learning_rate": 0.0002,
+      "loss": 1.8951,
+      "step": 1380
+    },
+    {
+      "epoch": 3.1519274376417235,
+      "grad_norm": 0.761993944644928,
+      "learning_rate": 0.0002,
+      "loss": 1.8308,
+      "step": 1390
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6993107795715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9513,
+      "step": 1400
+    },
+    {
+      "epoch": 3.197278911564626,
+      "grad_norm": 0.6398035287857056,
+      "learning_rate": 0.0002,
+      "loss": 1.8857,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2199546485260773,
+      "grad_norm": 0.8602580428123474,
+      "learning_rate": 0.0002,
+      "loss": 1.9041,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2426303854875282,
+      "grad_norm": 0.7799932360649109,
+      "learning_rate": 0.0002,
+      "loss": 1.9629,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2653061224489797,
+      "grad_norm": 0.6598666906356812,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 1440
+    },
+    {
+      "epoch": 3.287981859410431,
+      "grad_norm": 0.6513776779174805,
+      "learning_rate": 0.0002,
+      "loss": 1.8772,
+      "step": 1450
+    },
+    {
+      "epoch": 3.310657596371882,
+      "grad_norm": 0.7690088748931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8568,
+      "step": 1460
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 1.0070042610168457,
+      "learning_rate": 0.0002,
+      "loss": 1.8535,
+      "step": 1470
+    },
+    {
+      "epoch": 3.3560090702947845,
+      "grad_norm": 0.8606905341148376,
+      "learning_rate": 0.0002,
+      "loss": 1.8971,
+      "step": 1480
+    },
+    {
+      "epoch": 3.378684807256236,
+      "grad_norm": 0.7596806287765503,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4013605442176873,
+      "grad_norm": 0.6961908936500549,
+      "learning_rate": 0.0002,
+      "loss": 1.9278,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4240362811791383,
+      "grad_norm": 0.8368703126907349,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 1510
+    },
+    {
+      "epoch": 3.4467120181405897,
+      "grad_norm": 0.6655045747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.924,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4693877551020407,
+      "grad_norm": 0.7557174563407898,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1530
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.6713934540748596,
+      "learning_rate": 0.0002,
+      "loss": 1.8435,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5147392290249435,
+      "grad_norm": 0.7006916999816895,
+      "learning_rate": 0.0002,
+      "loss": 1.9016,
+      "step": 1550
+    },
+    {
+      "epoch": 3.5374149659863945,
+      "grad_norm": 0.6728846430778503,
+      "learning_rate": 0.0002,
+      "loss": 1.9208,
+      "step": 1560
+    },
+    {
+      "epoch": 3.560090702947846,
+      "grad_norm": 0.7479315400123596,
+      "learning_rate": 0.0002,
+      "loss": 1.8769,
+      "step": 1570
+    },
+    {
+      "epoch": 3.582766439909297,
+      "grad_norm": 0.7596876621246338,
+      "learning_rate": 0.0002,
+      "loss": 1.8558,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6054421768707483,
+      "grad_norm": 0.8948362469673157,
+      "learning_rate": 0.0002,
+      "loss": 1.876,
+      "step": 1590
+    },
+    {
+      "epoch": 3.6281179138321997,
+      "grad_norm": 0.6952481269836426,
+      "learning_rate": 0.0002,
+      "loss": 1.8552,
+      "step": 1600
+    },
+    {
+      "epoch": 3.6507936507936507,
+      "grad_norm": 0.6494414210319519,
+      "learning_rate": 0.0002,
+      "loss": 1.933,
+      "step": 1610
+    },
+    {
+      "epoch": 3.673469387755102,
+      "grad_norm": 0.7438889145851135,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1620
+    },
+    {
+      "epoch": 3.696145124716553,
+      "grad_norm": 0.7470757961273193,
+      "learning_rate": 0.0002,
+      "loss": 1.9441,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7188208616780045,
+      "grad_norm": 0.8727335333824158,
+      "learning_rate": 0.0002,
+      "loss": 1.8685,
+      "step": 1640
+    },
+    {
+      "epoch": 3.741496598639456,
+      "grad_norm": 0.7064452767372131,
+      "learning_rate": 0.0002,
+      "loss": 1.8831,
+      "step": 1650
+    },
+    {
+      "epoch": 3.764172335600907,
+      "grad_norm": 0.6958122849464417,
+      "learning_rate": 0.0002,
+      "loss": 1.9743,
+      "step": 1660
+    },
+    {
+      "epoch": 3.7868480725623583,
+      "grad_norm": 0.6751590371131897,
+      "learning_rate": 0.0002,
+      "loss": 1.927,
+      "step": 1670
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.7274497151374817,
+      "learning_rate": 0.0002,
+      "loss": 1.8868,
+      "step": 1680
+    },
+    {
+      "epoch": 3.8321995464852607,
+      "grad_norm": 0.7311708331108093,
+      "learning_rate": 0.0002,
+      "loss": 1.942,
+      "step": 1690
+    },
+    {
+      "epoch": 3.854875283446712,
+      "grad_norm": 0.6703173518180847,
+      "learning_rate": 0.0002,
+      "loss": 1.8134,
+      "step": 1700
+    },
+    {
+      "epoch": 3.877551020408163,
+      "grad_norm": 0.6263390779495239,
+      "learning_rate": 0.0002,
+      "loss": 1.8687,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9002267573696145,
+      "grad_norm": 0.744022011756897,
+      "learning_rate": 0.0002,
+      "loss": 1.9206,
+      "step": 1720
+    },
+    {
+      "epoch": 3.9229024943310655,
+      "grad_norm": 0.7496722340583801,
+      "learning_rate": 0.0002,
+      "loss": 1.8961,
+      "step": 1730
+    },
+    {
+      "epoch": 3.945578231292517,
+      "grad_norm": 0.6877635717391968,
+      "learning_rate": 0.0002,
+      "loss": 1.9212,
+      "step": 1740
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.711262047290802,
+      "learning_rate": 0.0002,
+      "loss": 1.8504,
+      "step": 1750
+    },
+    {
+      "epoch": 3.9909297052154193,
+      "grad_norm": 0.7452033758163452,
+      "learning_rate": 0.0002,
+      "loss": 1.9315,
+      "step": 1760
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.065720319747925,
+      "eval_runtime": 14.7103,
+      "eval_samples_per_second": 34.466,
+      "eval_steps_per_second": 4.351,
+      "step": 1764
+    },
+    {
+      "epoch": 4.01360544217687,
+      "grad_norm": 0.6913991570472717,
+      "learning_rate": 0.0002,
+      "loss": 1.8729,
+      "step": 1770
+    },
+    {
+      "epoch": 4.036281179138322,
+      "grad_norm": 0.7323354482650757,
+      "learning_rate": 0.0002,
+      "loss": 1.7617,
+      "step": 1780
+    },
+    {
+      "epoch": 4.058956916099773,
+      "grad_norm": 0.7691134214401245,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1790
+    },
+    {
+      "epoch": 4.081632653061225,
+      "grad_norm": 0.7871596813201904,
+      "learning_rate": 0.0002,
+      "loss": 1.8102,
+      "step": 1800
+    },
+    {
+      "epoch": 4.104308390022676,
+      "grad_norm": 0.7079440355300903,
+      "learning_rate": 0.0002,
+      "loss": 1.7455,
+      "step": 1810
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 0.7468329668045044,
+      "learning_rate": 0.0002,
+      "loss": 1.7693,
+      "step": 1820
+    },
+    {
+      "epoch": 4.149659863945578,
+      "grad_norm": 0.8017606139183044,
+      "learning_rate": 0.0002,
+      "loss": 1.772,
+      "step": 1830
+    },
+    {
+      "epoch": 4.172335600907029,
+      "grad_norm": 0.8591130375862122,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 1840
+    },
+    {
+      "epoch": 4.195011337868481,
+      "grad_norm": 0.7866499423980713,
+      "learning_rate": 0.0002,
+      "loss": 1.7356,
+      "step": 1850
+    },
+    {
+      "epoch": 4.217687074829932,
+      "grad_norm": 0.8457089066505432,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 1860
+    },
+    {
+      "epoch": 4.240362811791384,
+      "grad_norm": 0.8171738386154175,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 1870
+    },
+    {
+      "epoch": 4.263038548752834,
+      "grad_norm": 0.7377166748046875,
+      "learning_rate": 0.0002,
+      "loss": 1.7786,
+      "step": 1880
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.7925628423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.787,
+      "step": 1890
+    },
+    {
+      "epoch": 4.308390022675737,
+      "grad_norm": 1.0107476711273193,
+      "learning_rate": 0.0002,
+      "loss": 1.8634,
+      "step": 1900
+    },
+    {
+      "epoch": 4.331065759637188,
+      "grad_norm": 0.7791350483894348,
+      "learning_rate": 0.0002,
+      "loss": 1.8193,
+      "step": 1910
+    },
+    {
+      "epoch": 4.35374149659864,
+      "grad_norm": 0.7173832654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.8431,
+      "step": 1920
+    },
+    {
+      "epoch": 4.37641723356009,
+      "grad_norm": 0.8341949582099915,
+      "learning_rate": 0.0002,
+      "loss": 1.83,
+      "step": 1930
+    },
+    {
+      "epoch": 4.399092970521542,
+      "grad_norm": 0.7944530844688416,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 1940
+    },
+    {
+      "epoch": 4.421768707482993,
+      "grad_norm": 0.7409125566482544,
+      "learning_rate": 0.0002,
+      "loss": 1.8161,
+      "step": 1950
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.9393547773361206,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 1960
+    },
+    {
+      "epoch": 4.467120181405896,
+      "grad_norm": 0.8280852437019348,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 1970
+    },
+    {
+      "epoch": 4.489795918367347,
+      "grad_norm": 0.7249669432640076,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 4.512471655328798,
+      "grad_norm": 0.7199458479881287,
+      "learning_rate": 0.0002,
+      "loss": 1.7846,
+      "step": 1990
+    },
+    {
+      "epoch": 4.535147392290249,
+      "grad_norm": 0.8454286456108093,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 2000
+    },
+    {
+      "epoch": 4.557823129251701,
+      "grad_norm": 0.7305117249488831,
+      "learning_rate": 0.0002,
+      "loss": 1.8673,
+      "step": 2010
+    },
+    {
+      "epoch": 4.580498866213152,
+      "grad_norm": 0.7819753885269165,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 2020
+    },
+    {
+      "epoch": 4.603174603174603,
+      "grad_norm": 0.8010004162788391,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 2030
+    },
+    {
+      "epoch": 4.625850340136054,
+      "grad_norm": 0.7850886583328247,
+      "learning_rate": 0.0002,
+      "loss": 1.7921,
+      "step": 2040
+    },
+    {
+      "epoch": 4.648526077097506,
+      "grad_norm": 0.7433428168296814,
+      "learning_rate": 0.0002,
+      "loss": 1.9024,
+      "step": 2050
+    },
+    {
+      "epoch": 4.671201814058957,
+      "grad_norm": 0.7518399357795715,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 2060
+    },
+    {
+      "epoch": 4.6938775510204085,
+      "grad_norm": 0.7282178997993469,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 2070
+    },
+    {
+      "epoch": 4.71655328798186,
+      "grad_norm": 0.6950743794441223,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2080
+    },
+    {
+      "epoch": 4.73922902494331,
+      "grad_norm": 0.761830747127533,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 2090
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.8068088293075562,
+      "learning_rate": 0.0002,
+      "loss": 1.8254,
+      "step": 2100
+    },
+    {
+      "epoch": 4.784580498866213,
+      "grad_norm": 0.7580569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.8092,
+      "step": 2110
+    },
+    {
+      "epoch": 4.807256235827665,
+      "grad_norm": 0.9428760409355164,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2120
+    },
+    {
+      "epoch": 4.829931972789115,
+      "grad_norm": 0.7057452201843262,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 2130
+    },
+    {
+      "epoch": 4.852607709750567,
+      "grad_norm": 0.763630211353302,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 2140
+    },
+    {
+      "epoch": 4.875283446712018,
+      "grad_norm": 0.7298727035522461,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 2150
+    },
+    {
+      "epoch": 4.8979591836734695,
+      "grad_norm": 0.7376292943954468,
+      "learning_rate": 0.0002,
+      "loss": 1.8652,
+      "step": 2160
+    },
+    {
+      "epoch": 4.920634920634921,
+      "grad_norm": 0.8528493642807007,
+      "learning_rate": 0.0002,
+      "loss": 1.742,
+      "step": 2170
+    },
+    {
+      "epoch": 4.943310657596372,
+      "grad_norm": 0.7652511596679688,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2180
+    },
+    {
+      "epoch": 4.965986394557823,
+      "grad_norm": 0.7602494955062866,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 2190
+    },
+    {
+      "epoch": 4.988662131519274,
+      "grad_norm": 0.758914053440094,
+      "learning_rate": 0.0002,
+      "loss": 1.8278,
+      "step": 2200
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.083296298980713,
+      "eval_runtime": 20.2201,
+      "eval_samples_per_second": 25.074,
+      "eval_steps_per_second": 3.165,
+      "step": 2205
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 2.694950191890432e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2205/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f02f7067dad250ec6c2a1477a3103f93bdcc540d
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:840cfce3e80e39c50f59719cf91e382fb752e24e5a1c6cb8d44bd984a0019662
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6f625f95c221ca82dbbaf68cb25425e93ca47fc5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e5d25578ea19a1964f841d96d5478dd67c82f49bf8221581c3a9edcaff47f2a
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e646af610597efbdd511ed3310895edf8fd6b92d
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6dcf28783c7c05142319c5f409f2aee91b0443fa09563b9e472b607ace45e5a
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5784384b05ea93a82136e0d4f681dd48477739cd
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:550791043649ed75b1ff4f32159a98afbcc2337d75003352db8849d24012a186
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..4bf615e59ef0506f881f500e754e0e355abb66a8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/trainer_state.json
@@ -0,0 +1,1929 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 2646,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    },
+    {
+      "epoch": 3.015873015873016,
+      "grad_norm": 0.625606119632721,
+      "learning_rate": 0.0002,
+      "loss": 1.8798,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0385487528344672,
+      "grad_norm": 0.6679854989051819,
+      "learning_rate": 0.0002,
+      "loss": 1.8229,
+      "step": 1340
+    },
+    {
+      "epoch": 3.061224489795918,
+      "grad_norm": 0.6756007671356201,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1350
+    },
+    {
+      "epoch": 3.0839002267573696,
+      "grad_norm": 0.8019710779190063,
+      "learning_rate": 0.0002,
+      "loss": 1.8126,
+      "step": 1360
+    },
+    {
+      "epoch": 3.106575963718821,
+      "grad_norm": 0.9087499976158142,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1370
+    },
+    {
+      "epoch": 3.129251700680272,
+      "grad_norm": 0.7627860903739929,
+      "learning_rate": 0.0002,
+      "loss": 1.8951,
+      "step": 1380
+    },
+    {
+      "epoch": 3.1519274376417235,
+      "grad_norm": 0.761993944644928,
+      "learning_rate": 0.0002,
+      "loss": 1.8308,
+      "step": 1390
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6993107795715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9513,
+      "step": 1400
+    },
+    {
+      "epoch": 3.197278911564626,
+      "grad_norm": 0.6398035287857056,
+      "learning_rate": 0.0002,
+      "loss": 1.8857,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2199546485260773,
+      "grad_norm": 0.8602580428123474,
+      "learning_rate": 0.0002,
+      "loss": 1.9041,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2426303854875282,
+      "grad_norm": 0.7799932360649109,
+      "learning_rate": 0.0002,
+      "loss": 1.9629,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2653061224489797,
+      "grad_norm": 0.6598666906356812,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 1440
+    },
+    {
+      "epoch": 3.287981859410431,
+      "grad_norm": 0.6513776779174805,
+      "learning_rate": 0.0002,
+      "loss": 1.8772,
+      "step": 1450
+    },
+    {
+      "epoch": 3.310657596371882,
+      "grad_norm": 0.7690088748931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8568,
+      "step": 1460
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 1.0070042610168457,
+      "learning_rate": 0.0002,
+      "loss": 1.8535,
+      "step": 1470
+    },
+    {
+      "epoch": 3.3560090702947845,
+      "grad_norm": 0.8606905341148376,
+      "learning_rate": 0.0002,
+      "loss": 1.8971,
+      "step": 1480
+    },
+    {
+      "epoch": 3.378684807256236,
+      "grad_norm": 0.7596806287765503,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4013605442176873,
+      "grad_norm": 0.6961908936500549,
+      "learning_rate": 0.0002,
+      "loss": 1.9278,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4240362811791383,
+      "grad_norm": 0.8368703126907349,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 1510
+    },
+    {
+      "epoch": 3.4467120181405897,
+      "grad_norm": 0.6655045747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.924,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4693877551020407,
+      "grad_norm": 0.7557174563407898,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1530
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.6713934540748596,
+      "learning_rate": 0.0002,
+      "loss": 1.8435,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5147392290249435,
+      "grad_norm": 0.7006916999816895,
+      "learning_rate": 0.0002,
+      "loss": 1.9016,
+      "step": 1550
+    },
+    {
+      "epoch": 3.5374149659863945,
+      "grad_norm": 0.6728846430778503,
+      "learning_rate": 0.0002,
+      "loss": 1.9208,
+      "step": 1560
+    },
+    {
+      "epoch": 3.560090702947846,
+      "grad_norm": 0.7479315400123596,
+      "learning_rate": 0.0002,
+      "loss": 1.8769,
+      "step": 1570
+    },
+    {
+      "epoch": 3.582766439909297,
+      "grad_norm": 0.7596876621246338,
+      "learning_rate": 0.0002,
+      "loss": 1.8558,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6054421768707483,
+      "grad_norm": 0.8948362469673157,
+      "learning_rate": 0.0002,
+      "loss": 1.876,
+      "step": 1590
+    },
+    {
+      "epoch": 3.6281179138321997,
+      "grad_norm": 0.6952481269836426,
+      "learning_rate": 0.0002,
+      "loss": 1.8552,
+      "step": 1600
+    },
+    {
+      "epoch": 3.6507936507936507,
+      "grad_norm": 0.6494414210319519,
+      "learning_rate": 0.0002,
+      "loss": 1.933,
+      "step": 1610
+    },
+    {
+      "epoch": 3.673469387755102,
+      "grad_norm": 0.7438889145851135,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1620
+    },
+    {
+      "epoch": 3.696145124716553,
+      "grad_norm": 0.7470757961273193,
+      "learning_rate": 0.0002,
+      "loss": 1.9441,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7188208616780045,
+      "grad_norm": 0.8727335333824158,
+      "learning_rate": 0.0002,
+      "loss": 1.8685,
+      "step": 1640
+    },
+    {
+      "epoch": 3.741496598639456,
+      "grad_norm": 0.7064452767372131,
+      "learning_rate": 0.0002,
+      "loss": 1.8831,
+      "step": 1650
+    },
+    {
+      "epoch": 3.764172335600907,
+      "grad_norm": 0.6958122849464417,
+      "learning_rate": 0.0002,
+      "loss": 1.9743,
+      "step": 1660
+    },
+    {
+      "epoch": 3.7868480725623583,
+      "grad_norm": 0.6751590371131897,
+      "learning_rate": 0.0002,
+      "loss": 1.927,
+      "step": 1670
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.7274497151374817,
+      "learning_rate": 0.0002,
+      "loss": 1.8868,
+      "step": 1680
+    },
+    {
+      "epoch": 3.8321995464852607,
+      "grad_norm": 0.7311708331108093,
+      "learning_rate": 0.0002,
+      "loss": 1.942,
+      "step": 1690
+    },
+    {
+      "epoch": 3.854875283446712,
+      "grad_norm": 0.6703173518180847,
+      "learning_rate": 0.0002,
+      "loss": 1.8134,
+      "step": 1700
+    },
+    {
+      "epoch": 3.877551020408163,
+      "grad_norm": 0.6263390779495239,
+      "learning_rate": 0.0002,
+      "loss": 1.8687,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9002267573696145,
+      "grad_norm": 0.744022011756897,
+      "learning_rate": 0.0002,
+      "loss": 1.9206,
+      "step": 1720
+    },
+    {
+      "epoch": 3.9229024943310655,
+      "grad_norm": 0.7496722340583801,
+      "learning_rate": 0.0002,
+      "loss": 1.8961,
+      "step": 1730
+    },
+    {
+      "epoch": 3.945578231292517,
+      "grad_norm": 0.6877635717391968,
+      "learning_rate": 0.0002,
+      "loss": 1.9212,
+      "step": 1740
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.711262047290802,
+      "learning_rate": 0.0002,
+      "loss": 1.8504,
+      "step": 1750
+    },
+    {
+      "epoch": 3.9909297052154193,
+      "grad_norm": 0.7452033758163452,
+      "learning_rate": 0.0002,
+      "loss": 1.9315,
+      "step": 1760
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.065720319747925,
+      "eval_runtime": 14.7103,
+      "eval_samples_per_second": 34.466,
+      "eval_steps_per_second": 4.351,
+      "step": 1764
+    },
+    {
+      "epoch": 4.01360544217687,
+      "grad_norm": 0.6913991570472717,
+      "learning_rate": 0.0002,
+      "loss": 1.8729,
+      "step": 1770
+    },
+    {
+      "epoch": 4.036281179138322,
+      "grad_norm": 0.7323354482650757,
+      "learning_rate": 0.0002,
+      "loss": 1.7617,
+      "step": 1780
+    },
+    {
+      "epoch": 4.058956916099773,
+      "grad_norm": 0.7691134214401245,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1790
+    },
+    {
+      "epoch": 4.081632653061225,
+      "grad_norm": 0.7871596813201904,
+      "learning_rate": 0.0002,
+      "loss": 1.8102,
+      "step": 1800
+    },
+    {
+      "epoch": 4.104308390022676,
+      "grad_norm": 0.7079440355300903,
+      "learning_rate": 0.0002,
+      "loss": 1.7455,
+      "step": 1810
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 0.7468329668045044,
+      "learning_rate": 0.0002,
+      "loss": 1.7693,
+      "step": 1820
+    },
+    {
+      "epoch": 4.149659863945578,
+      "grad_norm": 0.8017606139183044,
+      "learning_rate": 0.0002,
+      "loss": 1.772,
+      "step": 1830
+    },
+    {
+      "epoch": 4.172335600907029,
+      "grad_norm": 0.8591130375862122,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 1840
+    },
+    {
+      "epoch": 4.195011337868481,
+      "grad_norm": 0.7866499423980713,
+      "learning_rate": 0.0002,
+      "loss": 1.7356,
+      "step": 1850
+    },
+    {
+      "epoch": 4.217687074829932,
+      "grad_norm": 0.8457089066505432,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 1860
+    },
+    {
+      "epoch": 4.240362811791384,
+      "grad_norm": 0.8171738386154175,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 1870
+    },
+    {
+      "epoch": 4.263038548752834,
+      "grad_norm": 0.7377166748046875,
+      "learning_rate": 0.0002,
+      "loss": 1.7786,
+      "step": 1880
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.7925628423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.787,
+      "step": 1890
+    },
+    {
+      "epoch": 4.308390022675737,
+      "grad_norm": 1.0107476711273193,
+      "learning_rate": 0.0002,
+      "loss": 1.8634,
+      "step": 1900
+    },
+    {
+      "epoch": 4.331065759637188,
+      "grad_norm": 0.7791350483894348,
+      "learning_rate": 0.0002,
+      "loss": 1.8193,
+      "step": 1910
+    },
+    {
+      "epoch": 4.35374149659864,
+      "grad_norm": 0.7173832654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.8431,
+      "step": 1920
+    },
+    {
+      "epoch": 4.37641723356009,
+      "grad_norm": 0.8341949582099915,
+      "learning_rate": 0.0002,
+      "loss": 1.83,
+      "step": 1930
+    },
+    {
+      "epoch": 4.399092970521542,
+      "grad_norm": 0.7944530844688416,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 1940
+    },
+    {
+      "epoch": 4.421768707482993,
+      "grad_norm": 0.7409125566482544,
+      "learning_rate": 0.0002,
+      "loss": 1.8161,
+      "step": 1950
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.9393547773361206,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 1960
+    },
+    {
+      "epoch": 4.467120181405896,
+      "grad_norm": 0.8280852437019348,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 1970
+    },
+    {
+      "epoch": 4.489795918367347,
+      "grad_norm": 0.7249669432640076,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 4.512471655328798,
+      "grad_norm": 0.7199458479881287,
+      "learning_rate": 0.0002,
+      "loss": 1.7846,
+      "step": 1990
+    },
+    {
+      "epoch": 4.535147392290249,
+      "grad_norm": 0.8454286456108093,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 2000
+    },
+    {
+      "epoch": 4.557823129251701,
+      "grad_norm": 0.7305117249488831,
+      "learning_rate": 0.0002,
+      "loss": 1.8673,
+      "step": 2010
+    },
+    {
+      "epoch": 4.580498866213152,
+      "grad_norm": 0.7819753885269165,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 2020
+    },
+    {
+      "epoch": 4.603174603174603,
+      "grad_norm": 0.8010004162788391,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 2030
+    },
+    {
+      "epoch": 4.625850340136054,
+      "grad_norm": 0.7850886583328247,
+      "learning_rate": 0.0002,
+      "loss": 1.7921,
+      "step": 2040
+    },
+    {
+      "epoch": 4.648526077097506,
+      "grad_norm": 0.7433428168296814,
+      "learning_rate": 0.0002,
+      "loss": 1.9024,
+      "step": 2050
+    },
+    {
+      "epoch": 4.671201814058957,
+      "grad_norm": 0.7518399357795715,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 2060
+    },
+    {
+      "epoch": 4.6938775510204085,
+      "grad_norm": 0.7282178997993469,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 2070
+    },
+    {
+      "epoch": 4.71655328798186,
+      "grad_norm": 0.6950743794441223,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2080
+    },
+    {
+      "epoch": 4.73922902494331,
+      "grad_norm": 0.761830747127533,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 2090
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.8068088293075562,
+      "learning_rate": 0.0002,
+      "loss": 1.8254,
+      "step": 2100
+    },
+    {
+      "epoch": 4.784580498866213,
+      "grad_norm": 0.7580569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.8092,
+      "step": 2110
+    },
+    {
+      "epoch": 4.807256235827665,
+      "grad_norm": 0.9428760409355164,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2120
+    },
+    {
+      "epoch": 4.829931972789115,
+      "grad_norm": 0.7057452201843262,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 2130
+    },
+    {
+      "epoch": 4.852607709750567,
+      "grad_norm": 0.763630211353302,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 2140
+    },
+    {
+      "epoch": 4.875283446712018,
+      "grad_norm": 0.7298727035522461,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 2150
+    },
+    {
+      "epoch": 4.8979591836734695,
+      "grad_norm": 0.7376292943954468,
+      "learning_rate": 0.0002,
+      "loss": 1.8652,
+      "step": 2160
+    },
+    {
+      "epoch": 4.920634920634921,
+      "grad_norm": 0.8528493642807007,
+      "learning_rate": 0.0002,
+      "loss": 1.742,
+      "step": 2170
+    },
+    {
+      "epoch": 4.943310657596372,
+      "grad_norm": 0.7652511596679688,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2180
+    },
+    {
+      "epoch": 4.965986394557823,
+      "grad_norm": 0.7602494955062866,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 2190
+    },
+    {
+      "epoch": 4.988662131519274,
+      "grad_norm": 0.758914053440094,
+      "learning_rate": 0.0002,
+      "loss": 1.8278,
+      "step": 2200
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.083296298980713,
+      "eval_runtime": 20.2201,
+      "eval_samples_per_second": 25.074,
+      "eval_steps_per_second": 3.165,
+      "step": 2205
+    },
+    {
+      "epoch": 5.011337868480726,
+      "grad_norm": 0.6990088224411011,
+      "learning_rate": 0.0002,
+      "loss": 1.7003,
+      "step": 2210
+    },
+    {
+      "epoch": 5.034013605442177,
+      "grad_norm": 0.7627242803573608,
+      "learning_rate": 0.0002,
+      "loss": 1.6451,
+      "step": 2220
+    },
+    {
+      "epoch": 5.0566893424036286,
+      "grad_norm": 0.8485700488090515,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 2230
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 0.8308766484260559,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 2240
+    },
+    {
+      "epoch": 5.1020408163265305,
+      "grad_norm": 0.825551450252533,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 2250
+    },
+    {
+      "epoch": 5.124716553287982,
+      "grad_norm": 0.8571329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2260
+    },
+    {
+      "epoch": 5.147392290249433,
+      "grad_norm": 0.7965090274810791,
+      "learning_rate": 0.0002,
+      "loss": 1.7304,
+      "step": 2270
+    },
+    {
+      "epoch": 5.170068027210885,
+      "grad_norm": 0.7926788330078125,
+      "learning_rate": 0.0002,
+      "loss": 1.713,
+      "step": 2280
+    },
+    {
+      "epoch": 5.192743764172335,
+      "grad_norm": 0.8400393128395081,
+      "learning_rate": 0.0002,
+      "loss": 1.7778,
+      "step": 2290
+    },
+    {
+      "epoch": 5.215419501133787,
+      "grad_norm": 0.9096018075942993,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2300
+    },
+    {
+      "epoch": 5.238095238095238,
+      "grad_norm": 0.8156160712242126,
+      "learning_rate": 0.0002,
+      "loss": 1.7158,
+      "step": 2310
+    },
+    {
+      "epoch": 5.26077097505669,
+      "grad_norm": 0.8418492674827576,
+      "learning_rate": 0.0002,
+      "loss": 1.6941,
+      "step": 2320
+    },
+    {
+      "epoch": 5.283446712018141,
+      "grad_norm": 0.7994900345802307,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2330
+    },
+    {
+      "epoch": 5.3061224489795915,
+      "grad_norm": 0.9292947053909302,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2340
+    },
+    {
+      "epoch": 5.328798185941043,
+      "grad_norm": 0.8896278142929077,
+      "learning_rate": 0.0002,
+      "loss": 1.8213,
+      "step": 2350
+    },
+    {
+      "epoch": 5.351473922902494,
+      "grad_norm": 0.9424751400947571,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2360
+    },
+    {
+      "epoch": 5.374149659863946,
+      "grad_norm": 0.8420370817184448,
+      "learning_rate": 0.0002,
+      "loss": 1.7032,
+      "step": 2370
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 0.8460527658462524,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 2380
+    },
+    {
+      "epoch": 5.419501133786848,
+      "grad_norm": 1.0171664953231812,
+      "learning_rate": 0.0002,
+      "loss": 1.6443,
+      "step": 2390
+    },
+    {
+      "epoch": 5.442176870748299,
+      "grad_norm": 0.9017615914344788,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2400
+    },
+    {
+      "epoch": 5.464852607709751,
+      "grad_norm": 0.8650022745132446,
+      "learning_rate": 0.0002,
+      "loss": 1.7046,
+      "step": 2410
+    },
+    {
+      "epoch": 5.487528344671202,
+      "grad_norm": 0.8252440094947815,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2420
+    },
+    {
+      "epoch": 5.510204081632653,
+      "grad_norm": 0.9326912760734558,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 2430
+    },
+    {
+      "epoch": 5.532879818594104,
+      "grad_norm": 0.8913490772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 2440
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.8588173985481262,
+      "learning_rate": 0.0002,
+      "loss": 1.7197,
+      "step": 2450
+    },
+    {
+      "epoch": 5.578231292517007,
+      "grad_norm": 0.8768706321716309,
+      "learning_rate": 0.0002,
+      "loss": 1.679,
+      "step": 2460
+    },
+    {
+      "epoch": 5.600907029478458,
+      "grad_norm": 0.9485137462615967,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2470
+    },
+    {
+      "epoch": 5.62358276643991,
+      "grad_norm": 0.8462285399436951,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 2480
+    },
+    {
+      "epoch": 5.646258503401361,
+      "grad_norm": 0.8390767574310303,
+      "learning_rate": 0.0002,
+      "loss": 1.7199,
+      "step": 2490
+    },
+    {
+      "epoch": 5.668934240362812,
+      "grad_norm": 0.9576658010482788,
+      "learning_rate": 0.0002,
+      "loss": 1.7819,
+      "step": 2500
+    },
+    {
+      "epoch": 5.691609977324263,
+      "grad_norm": 0.9528797268867493,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 2510
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.19835364818573,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 2520
+    },
+    {
+      "epoch": 5.736961451247166,
+      "grad_norm": 0.8822922110557556,
+      "learning_rate": 0.0002,
+      "loss": 1.6454,
+      "step": 2530
+    },
+    {
+      "epoch": 5.759637188208616,
+      "grad_norm": 0.9812421202659607,
+      "learning_rate": 0.0002,
+      "loss": 1.7769,
+      "step": 2540
+    },
+    {
+      "epoch": 5.782312925170068,
+      "grad_norm": 0.9178062677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 2550
+    },
+    {
+      "epoch": 5.804988662131519,
+      "grad_norm": 1.0217735767364502,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 2560
+    },
+    {
+      "epoch": 5.827664399092971,
+      "grad_norm": 0.8624402284622192,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2570
+    },
+    {
+      "epoch": 5.850340136054422,
+      "grad_norm": 0.8644461035728455,
+      "learning_rate": 0.0002,
+      "loss": 1.6976,
+      "step": 2580
+    },
+    {
+      "epoch": 5.8730158730158735,
+      "grad_norm": 0.8437298536300659,
+      "learning_rate": 0.0002,
+      "loss": 1.7498,
+      "step": 2590
+    },
+    {
+      "epoch": 5.895691609977324,
+      "grad_norm": 0.7952260971069336,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 2600
+    },
+    {
+      "epoch": 5.918367346938775,
+      "grad_norm": 0.9374001622200012,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 2610
+    },
+    {
+      "epoch": 5.941043083900227,
+      "grad_norm": 0.8157470226287842,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 2620
+    },
+    {
+      "epoch": 5.963718820861678,
+      "grad_norm": 0.8078457713127136,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 2630
+    },
+    {
+      "epoch": 5.986394557823129,
+      "grad_norm": 0.8926942348480225,
+      "learning_rate": 0.0002,
+      "loss": 1.7059,
+      "step": 2640
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.117249011993408,
+      "eval_runtime": 14.7114,
+      "eval_samples_per_second": 34.463,
+      "eval_steps_per_second": 4.35,
+      "step": 2646
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 3.2339402302685184e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-2646/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ba5e3a70b8cdf00d087266739ef92deceac6dd3f
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7caf3a89eadc20a728b8a590e558cbfa11813bc3532d8fa43087d1b5c296da7f
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..434b76953d74708a731ccc718a0cb752846ebbfd
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:293c3d6e83895447652a9be2597e0c75278cbbbec30ba3cde709cf08095118df
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0e29bb26d33655887b901d8592f96508f6483e6a
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef98ccc245162efcb771ef6dca6e89107ebdc19702d41662f45ac147f69fc16f
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8ab20168e1d0bb313bf9b2037fb084ec9a5f9490
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf07f40bad82bc4c8768de7621693647a66630cfe02662c69c8a8c63b374b71b
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5766a5115e3dc4e11bb229c15db578999c454893
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/trainer_state.json
@@ -0,0 +1,2245 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 7.0,
+  "eval_steps": 10,
+  "global_step": 3087,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    },
+    {
+      "epoch": 3.015873015873016,
+      "grad_norm": 0.625606119632721,
+      "learning_rate": 0.0002,
+      "loss": 1.8798,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0385487528344672,
+      "grad_norm": 0.6679854989051819,
+      "learning_rate": 0.0002,
+      "loss": 1.8229,
+      "step": 1340
+    },
+    {
+      "epoch": 3.061224489795918,
+      "grad_norm": 0.6756007671356201,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1350
+    },
+    {
+      "epoch": 3.0839002267573696,
+      "grad_norm": 0.8019710779190063,
+      "learning_rate": 0.0002,
+      "loss": 1.8126,
+      "step": 1360
+    },
+    {
+      "epoch": 3.106575963718821,
+      "grad_norm": 0.9087499976158142,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1370
+    },
+    {
+      "epoch": 3.129251700680272,
+      "grad_norm": 0.7627860903739929,
+      "learning_rate": 0.0002,
+      "loss": 1.8951,
+      "step": 1380
+    },
+    {
+      "epoch": 3.1519274376417235,
+      "grad_norm": 0.761993944644928,
+      "learning_rate": 0.0002,
+      "loss": 1.8308,
+      "step": 1390
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6993107795715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9513,
+      "step": 1400
+    },
+    {
+      "epoch": 3.197278911564626,
+      "grad_norm": 0.6398035287857056,
+      "learning_rate": 0.0002,
+      "loss": 1.8857,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2199546485260773,
+      "grad_norm": 0.8602580428123474,
+      "learning_rate": 0.0002,
+      "loss": 1.9041,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2426303854875282,
+      "grad_norm": 0.7799932360649109,
+      "learning_rate": 0.0002,
+      "loss": 1.9629,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2653061224489797,
+      "grad_norm": 0.6598666906356812,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 1440
+    },
+    {
+      "epoch": 3.287981859410431,
+      "grad_norm": 0.6513776779174805,
+      "learning_rate": 0.0002,
+      "loss": 1.8772,
+      "step": 1450
+    },
+    {
+      "epoch": 3.310657596371882,
+      "grad_norm": 0.7690088748931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8568,
+      "step": 1460
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 1.0070042610168457,
+      "learning_rate": 0.0002,
+      "loss": 1.8535,
+      "step": 1470
+    },
+    {
+      "epoch": 3.3560090702947845,
+      "grad_norm": 0.8606905341148376,
+      "learning_rate": 0.0002,
+      "loss": 1.8971,
+      "step": 1480
+    },
+    {
+      "epoch": 3.378684807256236,
+      "grad_norm": 0.7596806287765503,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4013605442176873,
+      "grad_norm": 0.6961908936500549,
+      "learning_rate": 0.0002,
+      "loss": 1.9278,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4240362811791383,
+      "grad_norm": 0.8368703126907349,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 1510
+    },
+    {
+      "epoch": 3.4467120181405897,
+      "grad_norm": 0.6655045747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.924,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4693877551020407,
+      "grad_norm": 0.7557174563407898,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1530
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.6713934540748596,
+      "learning_rate": 0.0002,
+      "loss": 1.8435,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5147392290249435,
+      "grad_norm": 0.7006916999816895,
+      "learning_rate": 0.0002,
+      "loss": 1.9016,
+      "step": 1550
+    },
+    {
+      "epoch": 3.5374149659863945,
+      "grad_norm": 0.6728846430778503,
+      "learning_rate": 0.0002,
+      "loss": 1.9208,
+      "step": 1560
+    },
+    {
+      "epoch": 3.560090702947846,
+      "grad_norm": 0.7479315400123596,
+      "learning_rate": 0.0002,
+      "loss": 1.8769,
+      "step": 1570
+    },
+    {
+      "epoch": 3.582766439909297,
+      "grad_norm": 0.7596876621246338,
+      "learning_rate": 0.0002,
+      "loss": 1.8558,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6054421768707483,
+      "grad_norm": 0.8948362469673157,
+      "learning_rate": 0.0002,
+      "loss": 1.876,
+      "step": 1590
+    },
+    {
+      "epoch": 3.6281179138321997,
+      "grad_norm": 0.6952481269836426,
+      "learning_rate": 0.0002,
+      "loss": 1.8552,
+      "step": 1600
+    },
+    {
+      "epoch": 3.6507936507936507,
+      "grad_norm": 0.6494414210319519,
+      "learning_rate": 0.0002,
+      "loss": 1.933,
+      "step": 1610
+    },
+    {
+      "epoch": 3.673469387755102,
+      "grad_norm": 0.7438889145851135,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1620
+    },
+    {
+      "epoch": 3.696145124716553,
+      "grad_norm": 0.7470757961273193,
+      "learning_rate": 0.0002,
+      "loss": 1.9441,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7188208616780045,
+      "grad_norm": 0.8727335333824158,
+      "learning_rate": 0.0002,
+      "loss": 1.8685,
+      "step": 1640
+    },
+    {
+      "epoch": 3.741496598639456,
+      "grad_norm": 0.7064452767372131,
+      "learning_rate": 0.0002,
+      "loss": 1.8831,
+      "step": 1650
+    },
+    {
+      "epoch": 3.764172335600907,
+      "grad_norm": 0.6958122849464417,
+      "learning_rate": 0.0002,
+      "loss": 1.9743,
+      "step": 1660
+    },
+    {
+      "epoch": 3.7868480725623583,
+      "grad_norm": 0.6751590371131897,
+      "learning_rate": 0.0002,
+      "loss": 1.927,
+      "step": 1670
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.7274497151374817,
+      "learning_rate": 0.0002,
+      "loss": 1.8868,
+      "step": 1680
+    },
+    {
+      "epoch": 3.8321995464852607,
+      "grad_norm": 0.7311708331108093,
+      "learning_rate": 0.0002,
+      "loss": 1.942,
+      "step": 1690
+    },
+    {
+      "epoch": 3.854875283446712,
+      "grad_norm": 0.6703173518180847,
+      "learning_rate": 0.0002,
+      "loss": 1.8134,
+      "step": 1700
+    },
+    {
+      "epoch": 3.877551020408163,
+      "grad_norm": 0.6263390779495239,
+      "learning_rate": 0.0002,
+      "loss": 1.8687,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9002267573696145,
+      "grad_norm": 0.744022011756897,
+      "learning_rate": 0.0002,
+      "loss": 1.9206,
+      "step": 1720
+    },
+    {
+      "epoch": 3.9229024943310655,
+      "grad_norm": 0.7496722340583801,
+      "learning_rate": 0.0002,
+      "loss": 1.8961,
+      "step": 1730
+    },
+    {
+      "epoch": 3.945578231292517,
+      "grad_norm": 0.6877635717391968,
+      "learning_rate": 0.0002,
+      "loss": 1.9212,
+      "step": 1740
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.711262047290802,
+      "learning_rate": 0.0002,
+      "loss": 1.8504,
+      "step": 1750
+    },
+    {
+      "epoch": 3.9909297052154193,
+      "grad_norm": 0.7452033758163452,
+      "learning_rate": 0.0002,
+      "loss": 1.9315,
+      "step": 1760
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.065720319747925,
+      "eval_runtime": 14.7103,
+      "eval_samples_per_second": 34.466,
+      "eval_steps_per_second": 4.351,
+      "step": 1764
+    },
+    {
+      "epoch": 4.01360544217687,
+      "grad_norm": 0.6913991570472717,
+      "learning_rate": 0.0002,
+      "loss": 1.8729,
+      "step": 1770
+    },
+    {
+      "epoch": 4.036281179138322,
+      "grad_norm": 0.7323354482650757,
+      "learning_rate": 0.0002,
+      "loss": 1.7617,
+      "step": 1780
+    },
+    {
+      "epoch": 4.058956916099773,
+      "grad_norm": 0.7691134214401245,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1790
+    },
+    {
+      "epoch": 4.081632653061225,
+      "grad_norm": 0.7871596813201904,
+      "learning_rate": 0.0002,
+      "loss": 1.8102,
+      "step": 1800
+    },
+    {
+      "epoch": 4.104308390022676,
+      "grad_norm": 0.7079440355300903,
+      "learning_rate": 0.0002,
+      "loss": 1.7455,
+      "step": 1810
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 0.7468329668045044,
+      "learning_rate": 0.0002,
+      "loss": 1.7693,
+      "step": 1820
+    },
+    {
+      "epoch": 4.149659863945578,
+      "grad_norm": 0.8017606139183044,
+      "learning_rate": 0.0002,
+      "loss": 1.772,
+      "step": 1830
+    },
+    {
+      "epoch": 4.172335600907029,
+      "grad_norm": 0.8591130375862122,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 1840
+    },
+    {
+      "epoch": 4.195011337868481,
+      "grad_norm": 0.7866499423980713,
+      "learning_rate": 0.0002,
+      "loss": 1.7356,
+      "step": 1850
+    },
+    {
+      "epoch": 4.217687074829932,
+      "grad_norm": 0.8457089066505432,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 1860
+    },
+    {
+      "epoch": 4.240362811791384,
+      "grad_norm": 0.8171738386154175,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 1870
+    },
+    {
+      "epoch": 4.263038548752834,
+      "grad_norm": 0.7377166748046875,
+      "learning_rate": 0.0002,
+      "loss": 1.7786,
+      "step": 1880
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.7925628423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.787,
+      "step": 1890
+    },
+    {
+      "epoch": 4.308390022675737,
+      "grad_norm": 1.0107476711273193,
+      "learning_rate": 0.0002,
+      "loss": 1.8634,
+      "step": 1900
+    },
+    {
+      "epoch": 4.331065759637188,
+      "grad_norm": 0.7791350483894348,
+      "learning_rate": 0.0002,
+      "loss": 1.8193,
+      "step": 1910
+    },
+    {
+      "epoch": 4.35374149659864,
+      "grad_norm": 0.7173832654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.8431,
+      "step": 1920
+    },
+    {
+      "epoch": 4.37641723356009,
+      "grad_norm": 0.8341949582099915,
+      "learning_rate": 0.0002,
+      "loss": 1.83,
+      "step": 1930
+    },
+    {
+      "epoch": 4.399092970521542,
+      "grad_norm": 0.7944530844688416,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 1940
+    },
+    {
+      "epoch": 4.421768707482993,
+      "grad_norm": 0.7409125566482544,
+      "learning_rate": 0.0002,
+      "loss": 1.8161,
+      "step": 1950
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.9393547773361206,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 1960
+    },
+    {
+      "epoch": 4.467120181405896,
+      "grad_norm": 0.8280852437019348,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 1970
+    },
+    {
+      "epoch": 4.489795918367347,
+      "grad_norm": 0.7249669432640076,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 4.512471655328798,
+      "grad_norm": 0.7199458479881287,
+      "learning_rate": 0.0002,
+      "loss": 1.7846,
+      "step": 1990
+    },
+    {
+      "epoch": 4.535147392290249,
+      "grad_norm": 0.8454286456108093,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 2000
+    },
+    {
+      "epoch": 4.557823129251701,
+      "grad_norm": 0.7305117249488831,
+      "learning_rate": 0.0002,
+      "loss": 1.8673,
+      "step": 2010
+    },
+    {
+      "epoch": 4.580498866213152,
+      "grad_norm": 0.7819753885269165,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 2020
+    },
+    {
+      "epoch": 4.603174603174603,
+      "grad_norm": 0.8010004162788391,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 2030
+    },
+    {
+      "epoch": 4.625850340136054,
+      "grad_norm": 0.7850886583328247,
+      "learning_rate": 0.0002,
+      "loss": 1.7921,
+      "step": 2040
+    },
+    {
+      "epoch": 4.648526077097506,
+      "grad_norm": 0.7433428168296814,
+      "learning_rate": 0.0002,
+      "loss": 1.9024,
+      "step": 2050
+    },
+    {
+      "epoch": 4.671201814058957,
+      "grad_norm": 0.7518399357795715,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 2060
+    },
+    {
+      "epoch": 4.6938775510204085,
+      "grad_norm": 0.7282178997993469,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 2070
+    },
+    {
+      "epoch": 4.71655328798186,
+      "grad_norm": 0.6950743794441223,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2080
+    },
+    {
+      "epoch": 4.73922902494331,
+      "grad_norm": 0.761830747127533,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 2090
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.8068088293075562,
+      "learning_rate": 0.0002,
+      "loss": 1.8254,
+      "step": 2100
+    },
+    {
+      "epoch": 4.784580498866213,
+      "grad_norm": 0.7580569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.8092,
+      "step": 2110
+    },
+    {
+      "epoch": 4.807256235827665,
+      "grad_norm": 0.9428760409355164,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2120
+    },
+    {
+      "epoch": 4.829931972789115,
+      "grad_norm": 0.7057452201843262,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 2130
+    },
+    {
+      "epoch": 4.852607709750567,
+      "grad_norm": 0.763630211353302,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 2140
+    },
+    {
+      "epoch": 4.875283446712018,
+      "grad_norm": 0.7298727035522461,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 2150
+    },
+    {
+      "epoch": 4.8979591836734695,
+      "grad_norm": 0.7376292943954468,
+      "learning_rate": 0.0002,
+      "loss": 1.8652,
+      "step": 2160
+    },
+    {
+      "epoch": 4.920634920634921,
+      "grad_norm": 0.8528493642807007,
+      "learning_rate": 0.0002,
+      "loss": 1.742,
+      "step": 2170
+    },
+    {
+      "epoch": 4.943310657596372,
+      "grad_norm": 0.7652511596679688,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2180
+    },
+    {
+      "epoch": 4.965986394557823,
+      "grad_norm": 0.7602494955062866,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 2190
+    },
+    {
+      "epoch": 4.988662131519274,
+      "grad_norm": 0.758914053440094,
+      "learning_rate": 0.0002,
+      "loss": 1.8278,
+      "step": 2200
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.083296298980713,
+      "eval_runtime": 20.2201,
+      "eval_samples_per_second": 25.074,
+      "eval_steps_per_second": 3.165,
+      "step": 2205
+    },
+    {
+      "epoch": 5.011337868480726,
+      "grad_norm": 0.6990088224411011,
+      "learning_rate": 0.0002,
+      "loss": 1.7003,
+      "step": 2210
+    },
+    {
+      "epoch": 5.034013605442177,
+      "grad_norm": 0.7627242803573608,
+      "learning_rate": 0.0002,
+      "loss": 1.6451,
+      "step": 2220
+    },
+    {
+      "epoch": 5.0566893424036286,
+      "grad_norm": 0.8485700488090515,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 2230
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 0.8308766484260559,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 2240
+    },
+    {
+      "epoch": 5.1020408163265305,
+      "grad_norm": 0.825551450252533,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 2250
+    },
+    {
+      "epoch": 5.124716553287982,
+      "grad_norm": 0.8571329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2260
+    },
+    {
+      "epoch": 5.147392290249433,
+      "grad_norm": 0.7965090274810791,
+      "learning_rate": 0.0002,
+      "loss": 1.7304,
+      "step": 2270
+    },
+    {
+      "epoch": 5.170068027210885,
+      "grad_norm": 0.7926788330078125,
+      "learning_rate": 0.0002,
+      "loss": 1.713,
+      "step": 2280
+    },
+    {
+      "epoch": 5.192743764172335,
+      "grad_norm": 0.8400393128395081,
+      "learning_rate": 0.0002,
+      "loss": 1.7778,
+      "step": 2290
+    },
+    {
+      "epoch": 5.215419501133787,
+      "grad_norm": 0.9096018075942993,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2300
+    },
+    {
+      "epoch": 5.238095238095238,
+      "grad_norm": 0.8156160712242126,
+      "learning_rate": 0.0002,
+      "loss": 1.7158,
+      "step": 2310
+    },
+    {
+      "epoch": 5.26077097505669,
+      "grad_norm": 0.8418492674827576,
+      "learning_rate": 0.0002,
+      "loss": 1.6941,
+      "step": 2320
+    },
+    {
+      "epoch": 5.283446712018141,
+      "grad_norm": 0.7994900345802307,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2330
+    },
+    {
+      "epoch": 5.3061224489795915,
+      "grad_norm": 0.9292947053909302,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2340
+    },
+    {
+      "epoch": 5.328798185941043,
+      "grad_norm": 0.8896278142929077,
+      "learning_rate": 0.0002,
+      "loss": 1.8213,
+      "step": 2350
+    },
+    {
+      "epoch": 5.351473922902494,
+      "grad_norm": 0.9424751400947571,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2360
+    },
+    {
+      "epoch": 5.374149659863946,
+      "grad_norm": 0.8420370817184448,
+      "learning_rate": 0.0002,
+      "loss": 1.7032,
+      "step": 2370
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 0.8460527658462524,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 2380
+    },
+    {
+      "epoch": 5.419501133786848,
+      "grad_norm": 1.0171664953231812,
+      "learning_rate": 0.0002,
+      "loss": 1.6443,
+      "step": 2390
+    },
+    {
+      "epoch": 5.442176870748299,
+      "grad_norm": 0.9017615914344788,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2400
+    },
+    {
+      "epoch": 5.464852607709751,
+      "grad_norm": 0.8650022745132446,
+      "learning_rate": 0.0002,
+      "loss": 1.7046,
+      "step": 2410
+    },
+    {
+      "epoch": 5.487528344671202,
+      "grad_norm": 0.8252440094947815,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2420
+    },
+    {
+      "epoch": 5.510204081632653,
+      "grad_norm": 0.9326912760734558,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 2430
+    },
+    {
+      "epoch": 5.532879818594104,
+      "grad_norm": 0.8913490772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 2440
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.8588173985481262,
+      "learning_rate": 0.0002,
+      "loss": 1.7197,
+      "step": 2450
+    },
+    {
+      "epoch": 5.578231292517007,
+      "grad_norm": 0.8768706321716309,
+      "learning_rate": 0.0002,
+      "loss": 1.679,
+      "step": 2460
+    },
+    {
+      "epoch": 5.600907029478458,
+      "grad_norm": 0.9485137462615967,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2470
+    },
+    {
+      "epoch": 5.62358276643991,
+      "grad_norm": 0.8462285399436951,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 2480
+    },
+    {
+      "epoch": 5.646258503401361,
+      "grad_norm": 0.8390767574310303,
+      "learning_rate": 0.0002,
+      "loss": 1.7199,
+      "step": 2490
+    },
+    {
+      "epoch": 5.668934240362812,
+      "grad_norm": 0.9576658010482788,
+      "learning_rate": 0.0002,
+      "loss": 1.7819,
+      "step": 2500
+    },
+    {
+      "epoch": 5.691609977324263,
+      "grad_norm": 0.9528797268867493,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 2510
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.19835364818573,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 2520
+    },
+    {
+      "epoch": 5.736961451247166,
+      "grad_norm": 0.8822922110557556,
+      "learning_rate": 0.0002,
+      "loss": 1.6454,
+      "step": 2530
+    },
+    {
+      "epoch": 5.759637188208616,
+      "grad_norm": 0.9812421202659607,
+      "learning_rate": 0.0002,
+      "loss": 1.7769,
+      "step": 2540
+    },
+    {
+      "epoch": 5.782312925170068,
+      "grad_norm": 0.9178062677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 2550
+    },
+    {
+      "epoch": 5.804988662131519,
+      "grad_norm": 1.0217735767364502,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 2560
+    },
+    {
+      "epoch": 5.827664399092971,
+      "grad_norm": 0.8624402284622192,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2570
+    },
+    {
+      "epoch": 5.850340136054422,
+      "grad_norm": 0.8644461035728455,
+      "learning_rate": 0.0002,
+      "loss": 1.6976,
+      "step": 2580
+    },
+    {
+      "epoch": 5.8730158730158735,
+      "grad_norm": 0.8437298536300659,
+      "learning_rate": 0.0002,
+      "loss": 1.7498,
+      "step": 2590
+    },
+    {
+      "epoch": 5.895691609977324,
+      "grad_norm": 0.7952260971069336,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 2600
+    },
+    {
+      "epoch": 5.918367346938775,
+      "grad_norm": 0.9374001622200012,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 2610
+    },
+    {
+      "epoch": 5.941043083900227,
+      "grad_norm": 0.8157470226287842,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 2620
+    },
+    {
+      "epoch": 5.963718820861678,
+      "grad_norm": 0.8078457713127136,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 2630
+    },
+    {
+      "epoch": 5.986394557823129,
+      "grad_norm": 0.8926942348480225,
+      "learning_rate": 0.0002,
+      "loss": 1.7059,
+      "step": 2640
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.117249011993408,
+      "eval_runtime": 14.7114,
+      "eval_samples_per_second": 34.463,
+      "eval_steps_per_second": 4.35,
+      "step": 2646
+    },
+    {
+      "epoch": 6.00907029478458,
+      "grad_norm": 0.866627037525177,
+      "learning_rate": 0.0002,
+      "loss": 1.6913,
+      "step": 2650
+    },
+    {
+      "epoch": 6.031746031746032,
+      "grad_norm": 0.9636926054954529,
+      "learning_rate": 0.0002,
+      "loss": 1.5567,
+      "step": 2660
+    },
+    {
+      "epoch": 6.054421768707483,
+      "grad_norm": 0.9994971752166748,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2670
+    },
+    {
+      "epoch": 6.0770975056689345,
+      "grad_norm": 0.886381983757019,
+      "learning_rate": 0.0002,
+      "loss": 1.5724,
+      "step": 2680
+    },
+    {
+      "epoch": 6.099773242630386,
+      "grad_norm": 0.8596521615982056,
+      "learning_rate": 0.0002,
+      "loss": 1.7039,
+      "step": 2690
+    },
+    {
+      "epoch": 6.122448979591836,
+      "grad_norm": 1.0034657716751099,
+      "learning_rate": 0.0002,
+      "loss": 1.6268,
+      "step": 2700
+    },
+    {
+      "epoch": 6.145124716553288,
+      "grad_norm": 0.8961828947067261,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2710
+    },
+    {
+      "epoch": 6.167800453514739,
+      "grad_norm": 0.9269746541976929,
+      "learning_rate": 0.0002,
+      "loss": 1.5924,
+      "step": 2720
+    },
+    {
+      "epoch": 6.190476190476191,
+      "grad_norm": 0.9434096217155457,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 2730
+    },
+    {
+      "epoch": 6.213151927437642,
+      "grad_norm": 0.9009348750114441,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2740
+    },
+    {
+      "epoch": 6.235827664399093,
+      "grad_norm": 1.0572991371154785,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 2750
+    },
+    {
+      "epoch": 6.258503401360544,
+      "grad_norm": 0.996005654335022,
+      "learning_rate": 0.0002,
+      "loss": 1.6675,
+      "step": 2760
+    },
+    {
+      "epoch": 6.2811791383219955,
+      "grad_norm": 0.8857560753822327,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 2770
+    },
+    {
+      "epoch": 6.303854875283447,
+      "grad_norm": 0.9528077840805054,
+      "learning_rate": 0.0002,
+      "loss": 1.6365,
+      "step": 2780
+    },
+    {
+      "epoch": 6.326530612244898,
+      "grad_norm": 0.8799851536750793,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 2790
+    },
+    {
+      "epoch": 6.349206349206349,
+      "grad_norm": 0.9826605916023254,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 2800
+    },
+    {
+      "epoch": 6.3718820861678,
+      "grad_norm": 1.0445730686187744,
+      "learning_rate": 0.0002,
+      "loss": 1.5672,
+      "step": 2810
+    },
+    {
+      "epoch": 6.394557823129252,
+      "grad_norm": 1.0388351678848267,
+      "learning_rate": 0.0002,
+      "loss": 1.6476,
+      "step": 2820
+    },
+    {
+      "epoch": 6.417233560090703,
+      "grad_norm": 0.9709566831588745,
+      "learning_rate": 0.0002,
+      "loss": 1.5719,
+      "step": 2830
+    },
+    {
+      "epoch": 6.4399092970521545,
+      "grad_norm": 0.9056107401847839,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2840
+    },
+    {
+      "epoch": 6.462585034013605,
+      "grad_norm": 1.0178978443145752,
+      "learning_rate": 0.0002,
+      "loss": 1.6482,
+      "step": 2850
+    },
+    {
+      "epoch": 6.4852607709750565,
+      "grad_norm": 0.9525032639503479,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 2860
+    },
+    {
+      "epoch": 6.507936507936508,
+      "grad_norm": 0.9033688306808472,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2870
+    },
+    {
+      "epoch": 6.530612244897959,
+      "grad_norm": 0.9108299016952515,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 2880
+    },
+    {
+      "epoch": 6.553287981859411,
+      "grad_norm": 0.961175262928009,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 2890
+    },
+    {
+      "epoch": 6.575963718820862,
+      "grad_norm": 0.8874754309654236,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 2900
+    },
+    {
+      "epoch": 6.598639455782313,
+      "grad_norm": 0.9972973465919495,
+      "learning_rate": 0.0002,
+      "loss": 1.6003,
+      "step": 2910
+    },
+    {
+      "epoch": 6.621315192743764,
+      "grad_norm": 0.9361436367034912,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2920
+    },
+    {
+      "epoch": 6.6439909297052155,
+      "grad_norm": 0.9379439353942871,
+      "learning_rate": 0.0002,
+      "loss": 1.6858,
+      "step": 2930
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.9896080493927002,
+      "learning_rate": 0.0002,
+      "loss": 1.6558,
+      "step": 2940
+    },
+    {
+      "epoch": 6.6893424036281175,
+      "grad_norm": 1.0048781633377075,
+      "learning_rate": 0.0002,
+      "loss": 1.6045,
+      "step": 2950
+    },
+    {
+      "epoch": 6.712018140589569,
+      "grad_norm": 0.9979572892189026,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2960
+    },
+    {
+      "epoch": 6.73469387755102,
+      "grad_norm": 0.9461016654968262,
+      "learning_rate": 0.0002,
+      "loss": 1.6663,
+      "step": 2970
+    },
+    {
+      "epoch": 6.757369614512472,
+      "grad_norm": 0.9460693597793579,
+      "learning_rate": 0.0002,
+      "loss": 1.6948,
+      "step": 2980
+    },
+    {
+      "epoch": 6.780045351473923,
+      "grad_norm": 0.9581215977668762,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 2990
+    },
+    {
+      "epoch": 6.802721088435375,
+      "grad_norm": 0.8998590111732483,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 3000
+    },
+    {
+      "epoch": 6.825396825396825,
+      "grad_norm": 0.9226139187812805,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 3010
+    },
+    {
+      "epoch": 6.8480725623582765,
+      "grad_norm": 0.9358452558517456,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 3020
+    },
+    {
+      "epoch": 6.870748299319728,
+      "grad_norm": 1.1139743328094482,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 3030
+    },
+    {
+      "epoch": 6.893424036281179,
+      "grad_norm": 0.9886892437934875,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3040
+    },
+    {
+      "epoch": 6.91609977324263,
+      "grad_norm": 0.885146975517273,
+      "learning_rate": 0.0002,
+      "loss": 1.6329,
+      "step": 3050
+    },
+    {
+      "epoch": 6.938775510204081,
+      "grad_norm": 0.9668041467666626,
+      "learning_rate": 0.0002,
+      "loss": 1.6473,
+      "step": 3060
+    },
+    {
+      "epoch": 6.961451247165533,
+      "grad_norm": 0.9532362222671509,
+      "learning_rate": 0.0002,
+      "loss": 1.5385,
+      "step": 3070
+    },
+    {
+      "epoch": 6.984126984126984,
+      "grad_norm": 1.0338832139968872,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.155296564102173,
+      "eval_runtime": 14.7454,
+      "eval_samples_per_second": 34.384,
+      "eval_steps_per_second": 4.34,
+      "step": 3087
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 3.772930268646605e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3087/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..735fd39e1b4e41f47079b0a4f518cb7ec378d490
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e3be997d67f22627f78b726361c46da3f7daa58c513df1a33b9d824428f8ac6
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..98ce8be718f49a29af1034c3a2466f4cb601698c
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a45b005a31d7596eaae867308769aa66773002cdf075984969fad49cf37899b
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..36de6382f4994897503090c9662ee2d72feb0ed1
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7162879088c93aa584240f151016a3cc124af3ff1e99b1d2d0e09470497fc6bd
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9ae51606d673cb30063b72865e51fb627f68cb66
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3066afa0b56998b1ecc486f95f90caf0d64ef6ec1449fc1da5ee8ffcdb912533
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6737d4aa6ef529dcfcee1f98a64c2c3ed40c31d4
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/trainer_state.json
@@ -0,0 +1,2561 @@
+{
+  "best_metric": 2.061427116394043,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323",
+  "epoch": 8.0,
+  "eval_steps": 10,
+  "global_step": 3528,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    },
+    {
+      "epoch": 2.018140589569161,
+      "grad_norm": 0.7971214056015015,
+      "learning_rate": 0.0002,
+      "loss": 2.015,
+      "step": 890
+    },
+    {
+      "epoch": 2.0408163265306123,
+      "grad_norm": 0.6144835948944092,
+      "learning_rate": 0.0002,
+      "loss": 1.982,
+      "step": 900
+    },
+    {
+      "epoch": 2.0634920634920633,
+      "grad_norm": 0.7324559688568115,
+      "learning_rate": 0.0002,
+      "loss": 1.9391,
+      "step": 910
+    },
+    {
+      "epoch": 2.0861678004535147,
+      "grad_norm": 0.6465244293212891,
+      "learning_rate": 0.0002,
+      "loss": 1.9066,
+      "step": 920
+    },
+    {
+      "epoch": 2.108843537414966,
+      "grad_norm": 0.7424528002738953,
+      "learning_rate": 0.0002,
+      "loss": 1.9893,
+      "step": 930
+    },
+    {
+      "epoch": 2.131519274376417,
+      "grad_norm": 0.6717734336853027,
+      "learning_rate": 0.0002,
+      "loss": 1.9193,
+      "step": 940
+    },
+    {
+      "epoch": 2.1541950113378685,
+      "grad_norm": 0.7873768210411072,
+      "learning_rate": 0.0002,
+      "loss": 1.9569,
+      "step": 950
+    },
+    {
+      "epoch": 2.17687074829932,
+      "grad_norm": 0.6752461194992065,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 960
+    },
+    {
+      "epoch": 2.199546485260771,
+      "grad_norm": 0.6919583082199097,
+      "learning_rate": 0.0002,
+      "loss": 2.0406,
+      "step": 970
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.6355509161949158,
+      "learning_rate": 0.0002,
+      "loss": 1.9281,
+      "step": 980
+    },
+    {
+      "epoch": 2.2448979591836733,
+      "grad_norm": 0.770007312297821,
+      "learning_rate": 0.0002,
+      "loss": 1.9953,
+      "step": 990
+    },
+    {
+      "epoch": 2.2675736961451247,
+      "grad_norm": 0.6216670870780945,
+      "learning_rate": 0.0002,
+      "loss": 1.9589,
+      "step": 1000
+    },
+    {
+      "epoch": 2.290249433106576,
+      "grad_norm": 0.6880573034286499,
+      "learning_rate": 0.0002,
+      "loss": 1.9247,
+      "step": 1010
+    },
+    {
+      "epoch": 2.312925170068027,
+      "grad_norm": 0.7346075177192688,
+      "learning_rate": 0.0002,
+      "loss": 1.9426,
+      "step": 1020
+    },
+    {
+      "epoch": 2.3356009070294785,
+      "grad_norm": 0.6651889681816101,
+      "learning_rate": 0.0002,
+      "loss": 1.9461,
+      "step": 1030
+    },
+    {
+      "epoch": 2.35827664399093,
+      "grad_norm": 0.7324103713035583,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 1040
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.7186149954795837,
+      "learning_rate": 0.0002,
+      "loss": 1.9142,
+      "step": 1050
+    },
+    {
+      "epoch": 2.4036281179138324,
+      "grad_norm": 0.6849178671836853,
+      "learning_rate": 0.0002,
+      "loss": 1.9501,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4263038548752833,
+      "grad_norm": 0.6259638667106628,
+      "learning_rate": 0.0002,
+      "loss": 1.889,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4489795918367347,
+      "grad_norm": 0.6821024417877197,
+      "learning_rate": 0.0002,
+      "loss": 1.9446,
+      "step": 1080
+    },
+    {
+      "epoch": 2.471655328798186,
+      "grad_norm": 0.7019884586334229,
+      "learning_rate": 0.0002,
+      "loss": 1.9398,
+      "step": 1090
+    },
+    {
+      "epoch": 2.494331065759637,
+      "grad_norm": 0.6368714570999146,
+      "learning_rate": 0.0002,
+      "loss": 1.9325,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5170068027210886,
+      "grad_norm": 0.6608313918113708,
+      "learning_rate": 0.0002,
+      "loss": 1.9773,
+      "step": 1110
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.6855525970458984,
+      "learning_rate": 0.0002,
+      "loss": 1.9337,
+      "step": 1120
+    },
+    {
+      "epoch": 2.562358276643991,
+      "grad_norm": 0.7279609441757202,
+      "learning_rate": 0.0002,
+      "loss": 1.9347,
+      "step": 1130
+    },
+    {
+      "epoch": 2.5850340136054424,
+      "grad_norm": 0.6539294123649597,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6077097505668934,
+      "grad_norm": 0.7035511136054993,
+      "learning_rate": 0.0002,
+      "loss": 2.0517,
+      "step": 1150
+    },
+    {
+      "epoch": 2.630385487528345,
+      "grad_norm": 0.7031794786453247,
+      "learning_rate": 0.0002,
+      "loss": 1.9955,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6530612244897958,
+      "grad_norm": 0.7587798833847046,
+      "learning_rate": 0.0002,
+      "loss": 1.9549,
+      "step": 1170
+    },
+    {
+      "epoch": 2.675736961451247,
+      "grad_norm": 0.707339882850647,
+      "learning_rate": 0.0002,
+      "loss": 2.0014,
+      "step": 1180
+    },
+    {
+      "epoch": 2.6984126984126986,
+      "grad_norm": 0.6633496880531311,
+      "learning_rate": 0.0002,
+      "loss": 1.955,
+      "step": 1190
+    },
+    {
+      "epoch": 2.7210884353741496,
+      "grad_norm": 0.9233061671257019,
+      "learning_rate": 0.0002,
+      "loss": 1.9434,
+      "step": 1200
+    },
+    {
+      "epoch": 2.743764172335601,
+      "grad_norm": 0.6396011710166931,
+      "learning_rate": 0.0002,
+      "loss": 1.9146,
+      "step": 1210
+    },
+    {
+      "epoch": 2.766439909297052,
+      "grad_norm": 0.6500924229621887,
+      "learning_rate": 0.0002,
+      "loss": 1.9934,
+      "step": 1220
+    },
+    {
+      "epoch": 2.7891156462585034,
+      "grad_norm": 0.6185927391052246,
+      "learning_rate": 0.0002,
+      "loss": 1.9197,
+      "step": 1230
+    },
+    {
+      "epoch": 2.811791383219955,
+      "grad_norm": 0.6737818717956543,
+      "learning_rate": 0.0002,
+      "loss": 1.9809,
+      "step": 1240
+    },
+    {
+      "epoch": 2.834467120181406,
+      "grad_norm": 0.8148701786994934,
+      "learning_rate": 0.0002,
+      "loss": 1.9508,
+      "step": 1250
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.7563779354095459,
+      "learning_rate": 0.0002,
+      "loss": 1.9848,
+      "step": 1260
+    },
+    {
+      "epoch": 2.879818594104308,
+      "grad_norm": 0.6679276823997498,
+      "learning_rate": 0.0002,
+      "loss": 1.9822,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9024943310657596,
+      "grad_norm": 0.7446274757385254,
+      "learning_rate": 0.0002,
+      "loss": 2.0609,
+      "step": 1280
+    },
+    {
+      "epoch": 2.925170068027211,
+      "grad_norm": 0.6530991196632385,
+      "learning_rate": 0.0002,
+      "loss": 1.9251,
+      "step": 1290
+    },
+    {
+      "epoch": 2.947845804988662,
+      "grad_norm": 0.6491292119026184,
+      "learning_rate": 0.0002,
+      "loss": 1.9555,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9705215419501134,
+      "grad_norm": 0.6711354851722717,
+      "learning_rate": 0.0002,
+      "loss": 1.913,
+      "step": 1310
+    },
+    {
+      "epoch": 2.9931972789115644,
+      "grad_norm": 0.6327161192893982,
+      "learning_rate": 0.0002,
+      "loss": 2.0005,
+      "step": 1320
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.061427116394043,
+      "eval_runtime": 19.6731,
+      "eval_samples_per_second": 25.771,
+      "eval_steps_per_second": 3.253,
+      "step": 1323
+    },
+    {
+      "epoch": 3.015873015873016,
+      "grad_norm": 0.625606119632721,
+      "learning_rate": 0.0002,
+      "loss": 1.8798,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0385487528344672,
+      "grad_norm": 0.6679854989051819,
+      "learning_rate": 0.0002,
+      "loss": 1.8229,
+      "step": 1340
+    },
+    {
+      "epoch": 3.061224489795918,
+      "grad_norm": 0.6756007671356201,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1350
+    },
+    {
+      "epoch": 3.0839002267573696,
+      "grad_norm": 0.8019710779190063,
+      "learning_rate": 0.0002,
+      "loss": 1.8126,
+      "step": 1360
+    },
+    {
+      "epoch": 3.106575963718821,
+      "grad_norm": 0.9087499976158142,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 1370
+    },
+    {
+      "epoch": 3.129251700680272,
+      "grad_norm": 0.7627860903739929,
+      "learning_rate": 0.0002,
+      "loss": 1.8951,
+      "step": 1380
+    },
+    {
+      "epoch": 3.1519274376417235,
+      "grad_norm": 0.761993944644928,
+      "learning_rate": 0.0002,
+      "loss": 1.8308,
+      "step": 1390
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6993107795715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9513,
+      "step": 1400
+    },
+    {
+      "epoch": 3.197278911564626,
+      "grad_norm": 0.6398035287857056,
+      "learning_rate": 0.0002,
+      "loss": 1.8857,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2199546485260773,
+      "grad_norm": 0.8602580428123474,
+      "learning_rate": 0.0002,
+      "loss": 1.9041,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2426303854875282,
+      "grad_norm": 0.7799932360649109,
+      "learning_rate": 0.0002,
+      "loss": 1.9629,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2653061224489797,
+      "grad_norm": 0.6598666906356812,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 1440
+    },
+    {
+      "epoch": 3.287981859410431,
+      "grad_norm": 0.6513776779174805,
+      "learning_rate": 0.0002,
+      "loss": 1.8772,
+      "step": 1450
+    },
+    {
+      "epoch": 3.310657596371882,
+      "grad_norm": 0.7690088748931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8568,
+      "step": 1460
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 1.0070042610168457,
+      "learning_rate": 0.0002,
+      "loss": 1.8535,
+      "step": 1470
+    },
+    {
+      "epoch": 3.3560090702947845,
+      "grad_norm": 0.8606905341148376,
+      "learning_rate": 0.0002,
+      "loss": 1.8971,
+      "step": 1480
+    },
+    {
+      "epoch": 3.378684807256236,
+      "grad_norm": 0.7596806287765503,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4013605442176873,
+      "grad_norm": 0.6961908936500549,
+      "learning_rate": 0.0002,
+      "loss": 1.9278,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4240362811791383,
+      "grad_norm": 0.8368703126907349,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 1510
+    },
+    {
+      "epoch": 3.4467120181405897,
+      "grad_norm": 0.6655045747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.924,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4693877551020407,
+      "grad_norm": 0.7557174563407898,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1530
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.6713934540748596,
+      "learning_rate": 0.0002,
+      "loss": 1.8435,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5147392290249435,
+      "grad_norm": 0.7006916999816895,
+      "learning_rate": 0.0002,
+      "loss": 1.9016,
+      "step": 1550
+    },
+    {
+      "epoch": 3.5374149659863945,
+      "grad_norm": 0.6728846430778503,
+      "learning_rate": 0.0002,
+      "loss": 1.9208,
+      "step": 1560
+    },
+    {
+      "epoch": 3.560090702947846,
+      "grad_norm": 0.7479315400123596,
+      "learning_rate": 0.0002,
+      "loss": 1.8769,
+      "step": 1570
+    },
+    {
+      "epoch": 3.582766439909297,
+      "grad_norm": 0.7596876621246338,
+      "learning_rate": 0.0002,
+      "loss": 1.8558,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6054421768707483,
+      "grad_norm": 0.8948362469673157,
+      "learning_rate": 0.0002,
+      "loss": 1.876,
+      "step": 1590
+    },
+    {
+      "epoch": 3.6281179138321997,
+      "grad_norm": 0.6952481269836426,
+      "learning_rate": 0.0002,
+      "loss": 1.8552,
+      "step": 1600
+    },
+    {
+      "epoch": 3.6507936507936507,
+      "grad_norm": 0.6494414210319519,
+      "learning_rate": 0.0002,
+      "loss": 1.933,
+      "step": 1610
+    },
+    {
+      "epoch": 3.673469387755102,
+      "grad_norm": 0.7438889145851135,
+      "learning_rate": 0.0002,
+      "loss": 1.8516,
+      "step": 1620
+    },
+    {
+      "epoch": 3.696145124716553,
+      "grad_norm": 0.7470757961273193,
+      "learning_rate": 0.0002,
+      "loss": 1.9441,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7188208616780045,
+      "grad_norm": 0.8727335333824158,
+      "learning_rate": 0.0002,
+      "loss": 1.8685,
+      "step": 1640
+    },
+    {
+      "epoch": 3.741496598639456,
+      "grad_norm": 0.7064452767372131,
+      "learning_rate": 0.0002,
+      "loss": 1.8831,
+      "step": 1650
+    },
+    {
+      "epoch": 3.764172335600907,
+      "grad_norm": 0.6958122849464417,
+      "learning_rate": 0.0002,
+      "loss": 1.9743,
+      "step": 1660
+    },
+    {
+      "epoch": 3.7868480725623583,
+      "grad_norm": 0.6751590371131897,
+      "learning_rate": 0.0002,
+      "loss": 1.927,
+      "step": 1670
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.7274497151374817,
+      "learning_rate": 0.0002,
+      "loss": 1.8868,
+      "step": 1680
+    },
+    {
+      "epoch": 3.8321995464852607,
+      "grad_norm": 0.7311708331108093,
+      "learning_rate": 0.0002,
+      "loss": 1.942,
+      "step": 1690
+    },
+    {
+      "epoch": 3.854875283446712,
+      "grad_norm": 0.6703173518180847,
+      "learning_rate": 0.0002,
+      "loss": 1.8134,
+      "step": 1700
+    },
+    {
+      "epoch": 3.877551020408163,
+      "grad_norm": 0.6263390779495239,
+      "learning_rate": 0.0002,
+      "loss": 1.8687,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9002267573696145,
+      "grad_norm": 0.744022011756897,
+      "learning_rate": 0.0002,
+      "loss": 1.9206,
+      "step": 1720
+    },
+    {
+      "epoch": 3.9229024943310655,
+      "grad_norm": 0.7496722340583801,
+      "learning_rate": 0.0002,
+      "loss": 1.8961,
+      "step": 1730
+    },
+    {
+      "epoch": 3.945578231292517,
+      "grad_norm": 0.6877635717391968,
+      "learning_rate": 0.0002,
+      "loss": 1.9212,
+      "step": 1740
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.711262047290802,
+      "learning_rate": 0.0002,
+      "loss": 1.8504,
+      "step": 1750
+    },
+    {
+      "epoch": 3.9909297052154193,
+      "grad_norm": 0.7452033758163452,
+      "learning_rate": 0.0002,
+      "loss": 1.9315,
+      "step": 1760
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.065720319747925,
+      "eval_runtime": 14.7103,
+      "eval_samples_per_second": 34.466,
+      "eval_steps_per_second": 4.351,
+      "step": 1764
+    },
+    {
+      "epoch": 4.01360544217687,
+      "grad_norm": 0.6913991570472717,
+      "learning_rate": 0.0002,
+      "loss": 1.8729,
+      "step": 1770
+    },
+    {
+      "epoch": 4.036281179138322,
+      "grad_norm": 0.7323354482650757,
+      "learning_rate": 0.0002,
+      "loss": 1.7617,
+      "step": 1780
+    },
+    {
+      "epoch": 4.058956916099773,
+      "grad_norm": 0.7691134214401245,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1790
+    },
+    {
+      "epoch": 4.081632653061225,
+      "grad_norm": 0.7871596813201904,
+      "learning_rate": 0.0002,
+      "loss": 1.8102,
+      "step": 1800
+    },
+    {
+      "epoch": 4.104308390022676,
+      "grad_norm": 0.7079440355300903,
+      "learning_rate": 0.0002,
+      "loss": 1.7455,
+      "step": 1810
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 0.7468329668045044,
+      "learning_rate": 0.0002,
+      "loss": 1.7693,
+      "step": 1820
+    },
+    {
+      "epoch": 4.149659863945578,
+      "grad_norm": 0.8017606139183044,
+      "learning_rate": 0.0002,
+      "loss": 1.772,
+      "step": 1830
+    },
+    {
+      "epoch": 4.172335600907029,
+      "grad_norm": 0.8591130375862122,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 1840
+    },
+    {
+      "epoch": 4.195011337868481,
+      "grad_norm": 0.7866499423980713,
+      "learning_rate": 0.0002,
+      "loss": 1.7356,
+      "step": 1850
+    },
+    {
+      "epoch": 4.217687074829932,
+      "grad_norm": 0.8457089066505432,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 1860
+    },
+    {
+      "epoch": 4.240362811791384,
+      "grad_norm": 0.8171738386154175,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 1870
+    },
+    {
+      "epoch": 4.263038548752834,
+      "grad_norm": 0.7377166748046875,
+      "learning_rate": 0.0002,
+      "loss": 1.7786,
+      "step": 1880
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.7925628423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.787,
+      "step": 1890
+    },
+    {
+      "epoch": 4.308390022675737,
+      "grad_norm": 1.0107476711273193,
+      "learning_rate": 0.0002,
+      "loss": 1.8634,
+      "step": 1900
+    },
+    {
+      "epoch": 4.331065759637188,
+      "grad_norm": 0.7791350483894348,
+      "learning_rate": 0.0002,
+      "loss": 1.8193,
+      "step": 1910
+    },
+    {
+      "epoch": 4.35374149659864,
+      "grad_norm": 0.7173832654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.8431,
+      "step": 1920
+    },
+    {
+      "epoch": 4.37641723356009,
+      "grad_norm": 0.8341949582099915,
+      "learning_rate": 0.0002,
+      "loss": 1.83,
+      "step": 1930
+    },
+    {
+      "epoch": 4.399092970521542,
+      "grad_norm": 0.7944530844688416,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 1940
+    },
+    {
+      "epoch": 4.421768707482993,
+      "grad_norm": 0.7409125566482544,
+      "learning_rate": 0.0002,
+      "loss": 1.8161,
+      "step": 1950
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.9393547773361206,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 1960
+    },
+    {
+      "epoch": 4.467120181405896,
+      "grad_norm": 0.8280852437019348,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 1970
+    },
+    {
+      "epoch": 4.489795918367347,
+      "grad_norm": 0.7249669432640076,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 4.512471655328798,
+      "grad_norm": 0.7199458479881287,
+      "learning_rate": 0.0002,
+      "loss": 1.7846,
+      "step": 1990
+    },
+    {
+      "epoch": 4.535147392290249,
+      "grad_norm": 0.8454286456108093,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 2000
+    },
+    {
+      "epoch": 4.557823129251701,
+      "grad_norm": 0.7305117249488831,
+      "learning_rate": 0.0002,
+      "loss": 1.8673,
+      "step": 2010
+    },
+    {
+      "epoch": 4.580498866213152,
+      "grad_norm": 0.7819753885269165,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 2020
+    },
+    {
+      "epoch": 4.603174603174603,
+      "grad_norm": 0.8010004162788391,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 2030
+    },
+    {
+      "epoch": 4.625850340136054,
+      "grad_norm": 0.7850886583328247,
+      "learning_rate": 0.0002,
+      "loss": 1.7921,
+      "step": 2040
+    },
+    {
+      "epoch": 4.648526077097506,
+      "grad_norm": 0.7433428168296814,
+      "learning_rate": 0.0002,
+      "loss": 1.9024,
+      "step": 2050
+    },
+    {
+      "epoch": 4.671201814058957,
+      "grad_norm": 0.7518399357795715,
+      "learning_rate": 0.0002,
+      "loss": 1.8155,
+      "step": 2060
+    },
+    {
+      "epoch": 4.6938775510204085,
+      "grad_norm": 0.7282178997993469,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 2070
+    },
+    {
+      "epoch": 4.71655328798186,
+      "grad_norm": 0.6950743794441223,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2080
+    },
+    {
+      "epoch": 4.73922902494331,
+      "grad_norm": 0.761830747127533,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 2090
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.8068088293075562,
+      "learning_rate": 0.0002,
+      "loss": 1.8254,
+      "step": 2100
+    },
+    {
+      "epoch": 4.784580498866213,
+      "grad_norm": 0.7580569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.8092,
+      "step": 2110
+    },
+    {
+      "epoch": 4.807256235827665,
+      "grad_norm": 0.9428760409355164,
+      "learning_rate": 0.0002,
+      "loss": 1.8144,
+      "step": 2120
+    },
+    {
+      "epoch": 4.829931972789115,
+      "grad_norm": 0.7057452201843262,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 2130
+    },
+    {
+      "epoch": 4.852607709750567,
+      "grad_norm": 0.763630211353302,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 2140
+    },
+    {
+      "epoch": 4.875283446712018,
+      "grad_norm": 0.7298727035522461,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 2150
+    },
+    {
+      "epoch": 4.8979591836734695,
+      "grad_norm": 0.7376292943954468,
+      "learning_rate": 0.0002,
+      "loss": 1.8652,
+      "step": 2160
+    },
+    {
+      "epoch": 4.920634920634921,
+      "grad_norm": 0.8528493642807007,
+      "learning_rate": 0.0002,
+      "loss": 1.742,
+      "step": 2170
+    },
+    {
+      "epoch": 4.943310657596372,
+      "grad_norm": 0.7652511596679688,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2180
+    },
+    {
+      "epoch": 4.965986394557823,
+      "grad_norm": 0.7602494955062866,
+      "learning_rate": 0.0002,
+      "loss": 1.7994,
+      "step": 2190
+    },
+    {
+      "epoch": 4.988662131519274,
+      "grad_norm": 0.758914053440094,
+      "learning_rate": 0.0002,
+      "loss": 1.8278,
+      "step": 2200
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.083296298980713,
+      "eval_runtime": 20.2201,
+      "eval_samples_per_second": 25.074,
+      "eval_steps_per_second": 3.165,
+      "step": 2205
+    },
+    {
+      "epoch": 5.011337868480726,
+      "grad_norm": 0.6990088224411011,
+      "learning_rate": 0.0002,
+      "loss": 1.7003,
+      "step": 2210
+    },
+    {
+      "epoch": 5.034013605442177,
+      "grad_norm": 0.7627242803573608,
+      "learning_rate": 0.0002,
+      "loss": 1.6451,
+      "step": 2220
+    },
+    {
+      "epoch": 5.0566893424036286,
+      "grad_norm": 0.8485700488090515,
+      "learning_rate": 0.0002,
+      "loss": 1.7713,
+      "step": 2230
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 0.8308766484260559,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 2240
+    },
+    {
+      "epoch": 5.1020408163265305,
+      "grad_norm": 0.825551450252533,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 2250
+    },
+    {
+      "epoch": 5.124716553287982,
+      "grad_norm": 0.8571329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2260
+    },
+    {
+      "epoch": 5.147392290249433,
+      "grad_norm": 0.7965090274810791,
+      "learning_rate": 0.0002,
+      "loss": 1.7304,
+      "step": 2270
+    },
+    {
+      "epoch": 5.170068027210885,
+      "grad_norm": 0.7926788330078125,
+      "learning_rate": 0.0002,
+      "loss": 1.713,
+      "step": 2280
+    },
+    {
+      "epoch": 5.192743764172335,
+      "grad_norm": 0.8400393128395081,
+      "learning_rate": 0.0002,
+      "loss": 1.7778,
+      "step": 2290
+    },
+    {
+      "epoch": 5.215419501133787,
+      "grad_norm": 0.9096018075942993,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2300
+    },
+    {
+      "epoch": 5.238095238095238,
+      "grad_norm": 0.8156160712242126,
+      "learning_rate": 0.0002,
+      "loss": 1.7158,
+      "step": 2310
+    },
+    {
+      "epoch": 5.26077097505669,
+      "grad_norm": 0.8418492674827576,
+      "learning_rate": 0.0002,
+      "loss": 1.6941,
+      "step": 2320
+    },
+    {
+      "epoch": 5.283446712018141,
+      "grad_norm": 0.7994900345802307,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2330
+    },
+    {
+      "epoch": 5.3061224489795915,
+      "grad_norm": 0.9292947053909302,
+      "learning_rate": 0.0002,
+      "loss": 1.6869,
+      "step": 2340
+    },
+    {
+      "epoch": 5.328798185941043,
+      "grad_norm": 0.8896278142929077,
+      "learning_rate": 0.0002,
+      "loss": 1.8213,
+      "step": 2350
+    },
+    {
+      "epoch": 5.351473922902494,
+      "grad_norm": 0.9424751400947571,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2360
+    },
+    {
+      "epoch": 5.374149659863946,
+      "grad_norm": 0.8420370817184448,
+      "learning_rate": 0.0002,
+      "loss": 1.7032,
+      "step": 2370
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 0.8460527658462524,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 2380
+    },
+    {
+      "epoch": 5.419501133786848,
+      "grad_norm": 1.0171664953231812,
+      "learning_rate": 0.0002,
+      "loss": 1.6443,
+      "step": 2390
+    },
+    {
+      "epoch": 5.442176870748299,
+      "grad_norm": 0.9017615914344788,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2400
+    },
+    {
+      "epoch": 5.464852607709751,
+      "grad_norm": 0.8650022745132446,
+      "learning_rate": 0.0002,
+      "loss": 1.7046,
+      "step": 2410
+    },
+    {
+      "epoch": 5.487528344671202,
+      "grad_norm": 0.8252440094947815,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2420
+    },
+    {
+      "epoch": 5.510204081632653,
+      "grad_norm": 0.9326912760734558,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 2430
+    },
+    {
+      "epoch": 5.532879818594104,
+      "grad_norm": 0.8913490772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 2440
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.8588173985481262,
+      "learning_rate": 0.0002,
+      "loss": 1.7197,
+      "step": 2450
+    },
+    {
+      "epoch": 5.578231292517007,
+      "grad_norm": 0.8768706321716309,
+      "learning_rate": 0.0002,
+      "loss": 1.679,
+      "step": 2460
+    },
+    {
+      "epoch": 5.600907029478458,
+      "grad_norm": 0.9485137462615967,
+      "learning_rate": 0.0002,
+      "loss": 1.7005,
+      "step": 2470
+    },
+    {
+      "epoch": 5.62358276643991,
+      "grad_norm": 0.8462285399436951,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 2480
+    },
+    {
+      "epoch": 5.646258503401361,
+      "grad_norm": 0.8390767574310303,
+      "learning_rate": 0.0002,
+      "loss": 1.7199,
+      "step": 2490
+    },
+    {
+      "epoch": 5.668934240362812,
+      "grad_norm": 0.9576658010482788,
+      "learning_rate": 0.0002,
+      "loss": 1.7819,
+      "step": 2500
+    },
+    {
+      "epoch": 5.691609977324263,
+      "grad_norm": 0.9528797268867493,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 2510
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.19835364818573,
+      "learning_rate": 0.0002,
+      "loss": 1.6769,
+      "step": 2520
+    },
+    {
+      "epoch": 5.736961451247166,
+      "grad_norm": 0.8822922110557556,
+      "learning_rate": 0.0002,
+      "loss": 1.6454,
+      "step": 2530
+    },
+    {
+      "epoch": 5.759637188208616,
+      "grad_norm": 0.9812421202659607,
+      "learning_rate": 0.0002,
+      "loss": 1.7769,
+      "step": 2540
+    },
+    {
+      "epoch": 5.782312925170068,
+      "grad_norm": 0.9178062677383423,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 2550
+    },
+    {
+      "epoch": 5.804988662131519,
+      "grad_norm": 1.0217735767364502,
+      "learning_rate": 0.0002,
+      "loss": 1.685,
+      "step": 2560
+    },
+    {
+      "epoch": 5.827664399092971,
+      "grad_norm": 0.8624402284622192,
+      "learning_rate": 0.0002,
+      "loss": 1.722,
+      "step": 2570
+    },
+    {
+      "epoch": 5.850340136054422,
+      "grad_norm": 0.8644461035728455,
+      "learning_rate": 0.0002,
+      "loss": 1.6976,
+      "step": 2580
+    },
+    {
+      "epoch": 5.8730158730158735,
+      "grad_norm": 0.8437298536300659,
+      "learning_rate": 0.0002,
+      "loss": 1.7498,
+      "step": 2590
+    },
+    {
+      "epoch": 5.895691609977324,
+      "grad_norm": 0.7952260971069336,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 2600
+    },
+    {
+      "epoch": 5.918367346938775,
+      "grad_norm": 0.9374001622200012,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 2610
+    },
+    {
+      "epoch": 5.941043083900227,
+      "grad_norm": 0.8157470226287842,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 2620
+    },
+    {
+      "epoch": 5.963718820861678,
+      "grad_norm": 0.8078457713127136,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 2630
+    },
+    {
+      "epoch": 5.986394557823129,
+      "grad_norm": 0.8926942348480225,
+      "learning_rate": 0.0002,
+      "loss": 1.7059,
+      "step": 2640
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.117249011993408,
+      "eval_runtime": 14.7114,
+      "eval_samples_per_second": 34.463,
+      "eval_steps_per_second": 4.35,
+      "step": 2646
+    },
+    {
+      "epoch": 6.00907029478458,
+      "grad_norm": 0.866627037525177,
+      "learning_rate": 0.0002,
+      "loss": 1.6913,
+      "step": 2650
+    },
+    {
+      "epoch": 6.031746031746032,
+      "grad_norm": 0.9636926054954529,
+      "learning_rate": 0.0002,
+      "loss": 1.5567,
+      "step": 2660
+    },
+    {
+      "epoch": 6.054421768707483,
+      "grad_norm": 0.9994971752166748,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2670
+    },
+    {
+      "epoch": 6.0770975056689345,
+      "grad_norm": 0.886381983757019,
+      "learning_rate": 0.0002,
+      "loss": 1.5724,
+      "step": 2680
+    },
+    {
+      "epoch": 6.099773242630386,
+      "grad_norm": 0.8596521615982056,
+      "learning_rate": 0.0002,
+      "loss": 1.7039,
+      "step": 2690
+    },
+    {
+      "epoch": 6.122448979591836,
+      "grad_norm": 1.0034657716751099,
+      "learning_rate": 0.0002,
+      "loss": 1.6268,
+      "step": 2700
+    },
+    {
+      "epoch": 6.145124716553288,
+      "grad_norm": 0.8961828947067261,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2710
+    },
+    {
+      "epoch": 6.167800453514739,
+      "grad_norm": 0.9269746541976929,
+      "learning_rate": 0.0002,
+      "loss": 1.5924,
+      "step": 2720
+    },
+    {
+      "epoch": 6.190476190476191,
+      "grad_norm": 0.9434096217155457,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 2730
+    },
+    {
+      "epoch": 6.213151927437642,
+      "grad_norm": 0.9009348750114441,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2740
+    },
+    {
+      "epoch": 6.235827664399093,
+      "grad_norm": 1.0572991371154785,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 2750
+    },
+    {
+      "epoch": 6.258503401360544,
+      "grad_norm": 0.996005654335022,
+      "learning_rate": 0.0002,
+      "loss": 1.6675,
+      "step": 2760
+    },
+    {
+      "epoch": 6.2811791383219955,
+      "grad_norm": 0.8857560753822327,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 2770
+    },
+    {
+      "epoch": 6.303854875283447,
+      "grad_norm": 0.9528077840805054,
+      "learning_rate": 0.0002,
+      "loss": 1.6365,
+      "step": 2780
+    },
+    {
+      "epoch": 6.326530612244898,
+      "grad_norm": 0.8799851536750793,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 2790
+    },
+    {
+      "epoch": 6.349206349206349,
+      "grad_norm": 0.9826605916023254,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 2800
+    },
+    {
+      "epoch": 6.3718820861678,
+      "grad_norm": 1.0445730686187744,
+      "learning_rate": 0.0002,
+      "loss": 1.5672,
+      "step": 2810
+    },
+    {
+      "epoch": 6.394557823129252,
+      "grad_norm": 1.0388351678848267,
+      "learning_rate": 0.0002,
+      "loss": 1.6476,
+      "step": 2820
+    },
+    {
+      "epoch": 6.417233560090703,
+      "grad_norm": 0.9709566831588745,
+      "learning_rate": 0.0002,
+      "loss": 1.5719,
+      "step": 2830
+    },
+    {
+      "epoch": 6.4399092970521545,
+      "grad_norm": 0.9056107401847839,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 2840
+    },
+    {
+      "epoch": 6.462585034013605,
+      "grad_norm": 1.0178978443145752,
+      "learning_rate": 0.0002,
+      "loss": 1.6482,
+      "step": 2850
+    },
+    {
+      "epoch": 6.4852607709750565,
+      "grad_norm": 0.9525032639503479,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 2860
+    },
+    {
+      "epoch": 6.507936507936508,
+      "grad_norm": 0.9033688306808472,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2870
+    },
+    {
+      "epoch": 6.530612244897959,
+      "grad_norm": 0.9108299016952515,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 2880
+    },
+    {
+      "epoch": 6.553287981859411,
+      "grad_norm": 0.961175262928009,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 2890
+    },
+    {
+      "epoch": 6.575963718820862,
+      "grad_norm": 0.8874754309654236,
+      "learning_rate": 0.0002,
+      "loss": 1.5573,
+      "step": 2900
+    },
+    {
+      "epoch": 6.598639455782313,
+      "grad_norm": 0.9972973465919495,
+      "learning_rate": 0.0002,
+      "loss": 1.6003,
+      "step": 2910
+    },
+    {
+      "epoch": 6.621315192743764,
+      "grad_norm": 0.9361436367034912,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 2920
+    },
+    {
+      "epoch": 6.6439909297052155,
+      "grad_norm": 0.9379439353942871,
+      "learning_rate": 0.0002,
+      "loss": 1.6858,
+      "step": 2930
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.9896080493927002,
+      "learning_rate": 0.0002,
+      "loss": 1.6558,
+      "step": 2940
+    },
+    {
+      "epoch": 6.6893424036281175,
+      "grad_norm": 1.0048781633377075,
+      "learning_rate": 0.0002,
+      "loss": 1.6045,
+      "step": 2950
+    },
+    {
+      "epoch": 6.712018140589569,
+      "grad_norm": 0.9979572892189026,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2960
+    },
+    {
+      "epoch": 6.73469387755102,
+      "grad_norm": 0.9461016654968262,
+      "learning_rate": 0.0002,
+      "loss": 1.6663,
+      "step": 2970
+    },
+    {
+      "epoch": 6.757369614512472,
+      "grad_norm": 0.9460693597793579,
+      "learning_rate": 0.0002,
+      "loss": 1.6948,
+      "step": 2980
+    },
+    {
+      "epoch": 6.780045351473923,
+      "grad_norm": 0.9581215977668762,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 2990
+    },
+    {
+      "epoch": 6.802721088435375,
+      "grad_norm": 0.8998590111732483,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 3000
+    },
+    {
+      "epoch": 6.825396825396825,
+      "grad_norm": 0.9226139187812805,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 3010
+    },
+    {
+      "epoch": 6.8480725623582765,
+      "grad_norm": 0.9358452558517456,
+      "learning_rate": 0.0002,
+      "loss": 1.6407,
+      "step": 3020
+    },
+    {
+      "epoch": 6.870748299319728,
+      "grad_norm": 1.1139743328094482,
+      "learning_rate": 0.0002,
+      "loss": 1.6654,
+      "step": 3030
+    },
+    {
+      "epoch": 6.893424036281179,
+      "grad_norm": 0.9886892437934875,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3040
+    },
+    {
+      "epoch": 6.91609977324263,
+      "grad_norm": 0.885146975517273,
+      "learning_rate": 0.0002,
+      "loss": 1.6329,
+      "step": 3050
+    },
+    {
+      "epoch": 6.938775510204081,
+      "grad_norm": 0.9668041467666626,
+      "learning_rate": 0.0002,
+      "loss": 1.6473,
+      "step": 3060
+    },
+    {
+      "epoch": 6.961451247165533,
+      "grad_norm": 0.9532362222671509,
+      "learning_rate": 0.0002,
+      "loss": 1.5385,
+      "step": 3070
+    },
+    {
+      "epoch": 6.984126984126984,
+      "grad_norm": 1.0338832139968872,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.155296564102173,
+      "eval_runtime": 14.7454,
+      "eval_samples_per_second": 34.384,
+      "eval_steps_per_second": 4.34,
+      "step": 3087
+    },
+    {
+      "epoch": 7.006802721088436,
+      "grad_norm": 0.979254424571991,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3090
+    },
+    {
+      "epoch": 7.029478458049887,
+      "grad_norm": 1.1125438213348389,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 3100
+    },
+    {
+      "epoch": 7.052154195011338,
+      "grad_norm": 1.080736517906189,
+      "learning_rate": 0.0002,
+      "loss": 1.5674,
+      "step": 3110
+    },
+    {
+      "epoch": 7.074829931972789,
+      "grad_norm": 1.0562000274658203,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 3120
+    },
+    {
+      "epoch": 7.09750566893424,
+      "grad_norm": 1.022085428237915,
+      "learning_rate": 0.0002,
+      "loss": 1.4841,
+      "step": 3130
+    },
+    {
+      "epoch": 7.120181405895692,
+      "grad_norm": 1.0203129053115845,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 3140
+    },
+    {
+      "epoch": 7.142857142857143,
+      "grad_norm": 1.0613057613372803,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3150
+    },
+    {
+      "epoch": 7.165532879818594,
+      "grad_norm": 1.2065532207489014,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 3160
+    },
+    {
+      "epoch": 7.188208616780045,
+      "grad_norm": 1.0489833354949951,
+      "learning_rate": 0.0002,
+      "loss": 1.5023,
+      "step": 3170
+    },
+    {
+      "epoch": 7.210884353741497,
+      "grad_norm": 1.1445655822753906,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 3180
+    },
+    {
+      "epoch": 7.233560090702948,
+      "grad_norm": 1.0830376148223877,
+      "learning_rate": 0.0002,
+      "loss": 1.5307,
+      "step": 3190
+    },
+    {
+      "epoch": 7.2562358276643995,
+      "grad_norm": 1.152576208114624,
+      "learning_rate": 0.0002,
+      "loss": 1.5298,
+      "step": 3200
+    },
+    {
+      "epoch": 7.27891156462585,
+      "grad_norm": 1.2006800174713135,
+      "learning_rate": 0.0002,
+      "loss": 1.4823,
+      "step": 3210
+    },
+    {
+      "epoch": 7.301587301587301,
+      "grad_norm": 1.0562821626663208,
+      "learning_rate": 0.0002,
+      "loss": 1.4567,
+      "step": 3220
+    },
+    {
+      "epoch": 7.324263038548753,
+      "grad_norm": 1.0565770864486694,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 3230
+    },
+    {
+      "epoch": 7.346938775510204,
+      "grad_norm": 1.0950746536254883,
+      "learning_rate": 0.0002,
+      "loss": 1.5353,
+      "step": 3240
+    },
+    {
+      "epoch": 7.369614512471656,
+      "grad_norm": 1.0843168497085571,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 3250
+    },
+    {
+      "epoch": 7.392290249433106,
+      "grad_norm": 1.1089458465576172,
+      "learning_rate": 0.0002,
+      "loss": 1.5202,
+      "step": 3260
+    },
+    {
+      "epoch": 7.414965986394558,
+      "grad_norm": 1.1431858539581299,
+      "learning_rate": 0.0002,
+      "loss": 1.5477,
+      "step": 3270
+    },
+    {
+      "epoch": 7.437641723356009,
+      "grad_norm": 1.1291835308074951,
+      "learning_rate": 0.0002,
+      "loss": 1.477,
+      "step": 3280
+    },
+    {
+      "epoch": 7.4603174603174605,
+      "grad_norm": 1.1871713399887085,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 3290
+    },
+    {
+      "epoch": 7.482993197278912,
+      "grad_norm": 1.0716718435287476,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 3300
+    },
+    {
+      "epoch": 7.505668934240363,
+      "grad_norm": 1.0762914419174194,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 3310
+    },
+    {
+      "epoch": 7.528344671201814,
+      "grad_norm": 1.0819021463394165,
+      "learning_rate": 0.0002,
+      "loss": 1.5113,
+      "step": 3320
+    },
+    {
+      "epoch": 7.551020408163265,
+      "grad_norm": 1.1713459491729736,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 3330
+    },
+    {
+      "epoch": 7.573696145124717,
+      "grad_norm": 1.0692853927612305,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3340
+    },
+    {
+      "epoch": 7.596371882086168,
+      "grad_norm": 1.2106385231018066,
+      "learning_rate": 0.0002,
+      "loss": 1.5046,
+      "step": 3350
+    },
+    {
+      "epoch": 7.619047619047619,
+      "grad_norm": 1.116910457611084,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 3360
+    },
+    {
+      "epoch": 7.64172335600907,
+      "grad_norm": 1.105789303779602,
+      "learning_rate": 0.0002,
+      "loss": 1.505,
+      "step": 3370
+    },
+    {
+      "epoch": 7.6643990929705215,
+      "grad_norm": 1.1199218034744263,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 3380
+    },
+    {
+      "epoch": 7.687074829931973,
+      "grad_norm": 1.056763768196106,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 3390
+    },
+    {
+      "epoch": 7.709750566893424,
+      "grad_norm": 1.0940561294555664,
+      "learning_rate": 0.0002,
+      "loss": 1.5497,
+      "step": 3400
+    },
+    {
+      "epoch": 7.732426303854876,
+      "grad_norm": 1.100599765777588,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 3410
+    },
+    {
+      "epoch": 7.755102040816326,
+      "grad_norm": 1.2683327198028564,
+      "learning_rate": 0.0002,
+      "loss": 1.5392,
+      "step": 3420
+    },
+    {
+      "epoch": 7.777777777777778,
+      "grad_norm": 1.140558123588562,
+      "learning_rate": 0.0002,
+      "loss": 1.5954,
+      "step": 3430
+    },
+    {
+      "epoch": 7.800453514739229,
+      "grad_norm": 1.096411943435669,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 3440
+    },
+    {
+      "epoch": 7.8231292517006805,
+      "grad_norm": 1.1843889951705933,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3450
+    },
+    {
+      "epoch": 7.845804988662131,
+      "grad_norm": 1.1010003089904785,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 3460
+    },
+    {
+      "epoch": 7.8684807256235825,
+      "grad_norm": 1.0672950744628906,
+      "learning_rate": 0.0002,
+      "loss": 1.5447,
+      "step": 3470
+    },
+    {
+      "epoch": 7.891156462585034,
+      "grad_norm": 1.0926506519317627,
+      "learning_rate": 0.0002,
+      "loss": 1.5641,
+      "step": 3480
+    },
+    {
+      "epoch": 7.913832199546485,
+      "grad_norm": 1.0815032720565796,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 3490
+    },
+    {
+      "epoch": 7.936507936507937,
+      "grad_norm": 1.2033658027648926,
+      "learning_rate": 0.0002,
+      "loss": 1.5018,
+      "step": 3500
+    },
+    {
+      "epoch": 7.959183673469388,
+      "grad_norm": 1.131916880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 3510
+    },
+    {
+      "epoch": 7.981859410430839,
+      "grad_norm": 1.0860273838043213,
+      "learning_rate": 0.0002,
+      "loss": 1.5814,
+      "step": 3520
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.226490020751953,
+      "eval_runtime": 14.7307,
+      "eval_samples_per_second": 34.418,
+      "eval_steps_per_second": 4.345,
+      "step": 3528
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.311920307024691e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-3528/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9b64adba4ce2670d3038b3fbbc5b4c655df4f115
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce86b538842f840848ef486b6e9b1914f943757ed56567bc9823d7cb08f912ba
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a7d3fa8fbc5322816e870450792d575d4b1fb2c0
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:522be20e18bda6ec6e408e9f05afa70c980a1636db906ea0838d2d3a6a1cbfe6
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b3cdc190f06f9a295befba12f602da933756f82c
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49395e2e99103d2e2634e22bde45e7bd3b104cb1a39b9377a54238190134b62c
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a9fb326b5c0e197fffeddea24499cc1a65a06ad0
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:782a55c868c9769b697ad73a8dd9d936e5f24496d6ca9d2dc3449af0d5fd9b78
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a85031566c6c7a0f5d3e5fe3983569a5884b3c75
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/trainer_state.json
@@ -0,0 +1,349 @@
+{
+  "best_metric": 2.114302158355713,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441",
+  "epoch": 1.0,
+  "eval_steps": 10,
+  "global_step": 441,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 5389900383780864.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/README.md b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e99bbcd43df1c19d98706c7e3be95c93844c5349
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_model.safetensors b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5769426b2935d1785d06838354cdfd5d3e69a079
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b3580fe564564ec60ae26c9eae589c591e660af292320091ebe7262f148fdb8
+size 29500848
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/optimizer.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c3fdf0f8c196ee21c78f1b9909897ad4ba8715f8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0832cdc013f26202782fdc3b7bdbc56c68df3e2b04e424762883014286167524
+size 15064314
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/rng_state.pth b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b03762fc666c1dc67a64cf6943ddbd182c38963d
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4af14da6d6a0259551d6058adfcb7d38bd0e33e4826799837628a1240c29ddcd
+size 14244
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/scheduler.pt b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2daa1dd53f4bc7eb73efae365c3d54239b0ed968
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98328ad65ad5852018d577817b21ad3aaa3078612066fd00fd46e7e196f4095a
+size 1064
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/trainer_state.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8f7f6b17286382b499feee668737e656178e0275
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/trainer_state.json
@@ -0,0 +1,665 @@
+{
+  "best_metric": 2.073181629180908,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 882,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022675736961451247,
+      "grad_norm": 1.8457136154174805,
+      "learning_rate": 0.0002,
+      "loss": 4.7982,
+      "step": 10
+    },
+    {
+      "epoch": 0.045351473922902494,
+      "grad_norm": 1.0967960357666016,
+      "learning_rate": 0.0002,
+      "loss": 3.71,
+      "step": 20
+    },
+    {
+      "epoch": 0.06802721088435375,
+      "grad_norm": 0.8257759213447571,
+      "learning_rate": 0.0002,
+      "loss": 3.3086,
+      "step": 30
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 0.8248435258865356,
+      "learning_rate": 0.0002,
+      "loss": 2.923,
+      "step": 40
+    },
+    {
+      "epoch": 0.11337868480725624,
+      "grad_norm": 1.1190264225006104,
+      "learning_rate": 0.0002,
+      "loss": 2.8263,
+      "step": 50
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 1.273951768875122,
+      "learning_rate": 0.0002,
+      "loss": 2.6239,
+      "step": 60
+    },
+    {
+      "epoch": 0.15873015873015872,
+      "grad_norm": 1.0004316568374634,
+      "learning_rate": 0.0002,
+      "loss": 2.5203,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 0.8296012878417969,
+      "learning_rate": 0.0002,
+      "loss": 2.5055,
+      "step": 80
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 1.1563128232955933,
+      "learning_rate": 0.0002,
+      "loss": 2.5255,
+      "step": 90
+    },
+    {
+      "epoch": 0.22675736961451248,
+      "grad_norm": 0.7863451838493347,
+      "learning_rate": 0.0002,
+      "loss": 2.4731,
+      "step": 100
+    },
+    {
+      "epoch": 0.2494331065759637,
+      "grad_norm": 0.9167478680610657,
+      "learning_rate": 0.0002,
+      "loss": 2.3688,
+      "step": 110
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 1.104406476020813,
+      "learning_rate": 0.0002,
+      "loss": 2.336,
+      "step": 120
+    },
+    {
+      "epoch": 0.2947845804988662,
+      "grad_norm": 1.109564185142517,
+      "learning_rate": 0.0002,
+      "loss": 2.3063,
+      "step": 130
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.7416750192642212,
+      "learning_rate": 0.0002,
+      "loss": 2.23,
+      "step": 140
+    },
+    {
+      "epoch": 0.3401360544217687,
+      "grad_norm": 1.1499258279800415,
+      "learning_rate": 0.0002,
+      "loss": 2.3346,
+      "step": 150
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 0.9961133599281311,
+      "learning_rate": 0.0002,
+      "loss": 2.2853,
+      "step": 160
+    },
+    {
+      "epoch": 0.3854875283446712,
+      "grad_norm": 0.9257232546806335,
+      "learning_rate": 0.0002,
+      "loss": 2.2731,
+      "step": 170
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 1.098513126373291,
+      "learning_rate": 0.0002,
+      "loss": 2.1986,
+      "step": 180
+    },
+    {
+      "epoch": 0.4308390022675737,
+      "grad_norm": 0.8935237526893616,
+      "learning_rate": 0.0002,
+      "loss": 2.1685,
+      "step": 190
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 0.9654302597045898,
+      "learning_rate": 0.0002,
+      "loss": 2.2077,
+      "step": 200
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.9273215532302856,
+      "learning_rate": 0.0002,
+      "loss": 2.2421,
+      "step": 210
+    },
+    {
+      "epoch": 0.4988662131519274,
+      "grad_norm": 0.7222742438316345,
+      "learning_rate": 0.0002,
+      "loss": 2.1973,
+      "step": 220
+    },
+    {
+      "epoch": 0.5215419501133787,
+      "grad_norm": 0.7825018763542175,
+      "learning_rate": 0.0002,
+      "loss": 2.0619,
+      "step": 230
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 0.7758685350418091,
+      "learning_rate": 0.0002,
+      "loss": 2.2247,
+      "step": 240
+    },
+    {
+      "epoch": 0.5668934240362812,
+      "grad_norm": 0.916719913482666,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 250
+    },
+    {
+      "epoch": 0.5895691609977324,
+      "grad_norm": 0.696332573890686,
+      "learning_rate": 0.0002,
+      "loss": 2.0614,
+      "step": 260
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.7362682223320007,
+      "learning_rate": 0.0002,
+      "loss": 2.0763,
+      "step": 270
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.8184517025947571,
+      "learning_rate": 0.0002,
+      "loss": 2.223,
+      "step": 280
+    },
+    {
+      "epoch": 0.6575963718820862,
+      "grad_norm": 0.7944937348365784,
+      "learning_rate": 0.0002,
+      "loss": 2.1634,
+      "step": 290
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 0.7258933186531067,
+      "learning_rate": 0.0002,
+      "loss": 2.0922,
+      "step": 300
+    },
+    {
+      "epoch": 0.7029478458049887,
+      "grad_norm": 0.7037094831466675,
+      "learning_rate": 0.0002,
+      "loss": 2.1221,
+      "step": 310
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 0.7499093413352966,
+      "learning_rate": 0.0002,
+      "loss": 2.1132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7482993197278912,
+      "grad_norm": 0.7725803256034851,
+      "learning_rate": 0.0002,
+      "loss": 2.2321,
+      "step": 330
+    },
+    {
+      "epoch": 0.7709750566893424,
+      "grad_norm": 0.6771686673164368,
+      "learning_rate": 0.0002,
+      "loss": 2.1611,
+      "step": 340
+    },
+    {
+      "epoch": 0.7936507936507936,
+      "grad_norm": 0.7557646036148071,
+      "learning_rate": 0.0002,
+      "loss": 2.0693,
+      "step": 350
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.7430537939071655,
+      "learning_rate": 0.0002,
+      "loss": 2.0629,
+      "step": 360
+    },
+    {
+      "epoch": 0.8390022675736961,
+      "grad_norm": 0.8602280616760254,
+      "learning_rate": 0.0002,
+      "loss": 2.0731,
+      "step": 370
+    },
+    {
+      "epoch": 0.8616780045351474,
+      "grad_norm": 0.753440260887146,
+      "learning_rate": 0.0002,
+      "loss": 2.1836,
+      "step": 380
+    },
+    {
+      "epoch": 0.8843537414965986,
+      "grad_norm": 0.7217596769332886,
+      "learning_rate": 0.0002,
+      "loss": 2.1186,
+      "step": 390
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 0.6116304397583008,
+      "learning_rate": 0.0002,
+      "loss": 2.0223,
+      "step": 400
+    },
+    {
+      "epoch": 0.9297052154195011,
+      "grad_norm": 0.6829515695571899,
+      "learning_rate": 0.0002,
+      "loss": 2.0491,
+      "step": 410
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.733259916305542,
+      "learning_rate": 0.0002,
+      "loss": 2.1518,
+      "step": 420
+    },
+    {
+      "epoch": 0.9750566893424036,
+      "grad_norm": 0.6036455035209656,
+      "learning_rate": 0.0002,
+      "loss": 2.076,
+      "step": 430
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 0.8478217720985413,
+      "learning_rate": 0.0002,
+      "loss": 2.1526,
+      "step": 440
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.114302158355713,
+      "eval_runtime": 16.5108,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 3.876,
+      "step": 441
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 0.6996896266937256,
+      "learning_rate": 0.0002,
+      "loss": 1.9967,
+      "step": 450
+    },
+    {
+      "epoch": 1.0430839002267573,
+      "grad_norm": 0.7616267800331116,
+      "learning_rate": 0.0002,
+      "loss": 2.0471,
+      "step": 460
+    },
+    {
+      "epoch": 1.0657596371882085,
+      "grad_norm": 0.5978519320487976,
+      "learning_rate": 0.0002,
+      "loss": 2.0674,
+      "step": 470
+    },
+    {
+      "epoch": 1.08843537414966,
+      "grad_norm": 0.703987717628479,
+      "learning_rate": 0.0002,
+      "loss": 2.0803,
+      "step": 480
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.6707539558410645,
+      "learning_rate": 0.0002,
+      "loss": 2.0215,
+      "step": 490
+    },
+    {
+      "epoch": 1.1337868480725624,
+      "grad_norm": 0.6343862414360046,
+      "learning_rate": 0.0002,
+      "loss": 2.045,
+      "step": 500
+    },
+    {
+      "epoch": 1.1564625850340136,
+      "grad_norm": 0.8836050033569336,
+      "learning_rate": 0.0002,
+      "loss": 1.9753,
+      "step": 510
+    },
+    {
+      "epoch": 1.179138321995465,
+      "grad_norm": 0.6288692951202393,
+      "learning_rate": 0.0002,
+      "loss": 1.9537,
+      "step": 520
+    },
+    {
+      "epoch": 1.2018140589569162,
+      "grad_norm": 0.6671382784843445,
+      "learning_rate": 0.0002,
+      "loss": 2.0072,
+      "step": 530
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 0.7260547876358032,
+      "learning_rate": 0.0002,
+      "loss": 2.0631,
+      "step": 540
+    },
+    {
+      "epoch": 1.2471655328798186,
+      "grad_norm": 0.6919140219688416,
+      "learning_rate": 0.0002,
+      "loss": 2.1395,
+      "step": 550
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.6895381212234497,
+      "learning_rate": 0.0002,
+      "loss": 2.1162,
+      "step": 560
+    },
+    {
+      "epoch": 1.2925170068027212,
+      "grad_norm": 0.7131213545799255,
+      "learning_rate": 0.0002,
+      "loss": 2.1023,
+      "step": 570
+    },
+    {
+      "epoch": 1.3151927437641724,
+      "grad_norm": 0.6087185144424438,
+      "learning_rate": 0.0002,
+      "loss": 2.0381,
+      "step": 580
+    },
+    {
+      "epoch": 1.3378684807256236,
+      "grad_norm": 0.6506251096725464,
+      "learning_rate": 0.0002,
+      "loss": 2.054,
+      "step": 590
+    },
+    {
+      "epoch": 1.3605442176870748,
+      "grad_norm": 0.6999365091323853,
+      "learning_rate": 0.0002,
+      "loss": 2.0669,
+      "step": 600
+    },
+    {
+      "epoch": 1.383219954648526,
+      "grad_norm": 0.6305016875267029,
+      "learning_rate": 0.0002,
+      "loss": 2.0121,
+      "step": 610
+    },
+    {
+      "epoch": 1.4058956916099774,
+      "grad_norm": 0.6413465142250061,
+      "learning_rate": 0.0002,
+      "loss": 1.9727,
+      "step": 620
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.7571105360984802,
+      "learning_rate": 0.0002,
+      "loss": 2.0676,
+      "step": 630
+    },
+    {
+      "epoch": 1.4512471655328798,
+      "grad_norm": 0.8087857365608215,
+      "learning_rate": 0.0002,
+      "loss": 2.0838,
+      "step": 640
+    },
+    {
+      "epoch": 1.473922902494331,
+      "grad_norm": 0.6284025311470032,
+      "learning_rate": 0.0002,
+      "loss": 1.9932,
+      "step": 650
+    },
+    {
+      "epoch": 1.4965986394557822,
+      "grad_norm": 0.6630339026451111,
+      "learning_rate": 0.0002,
+      "loss": 2.0456,
+      "step": 660
+    },
+    {
+      "epoch": 1.5192743764172336,
+      "grad_norm": 0.6777389645576477,
+      "learning_rate": 0.0002,
+      "loss": 2.0778,
+      "step": 670
+    },
+    {
+      "epoch": 1.5419501133786848,
+      "grad_norm": 0.6965354681015015,
+      "learning_rate": 0.0002,
+      "loss": 2.0437,
+      "step": 680
+    },
+    {
+      "epoch": 1.564625850340136,
+      "grad_norm": 0.7124104499816895,
+      "learning_rate": 0.0002,
+      "loss": 2.056,
+      "step": 690
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.6652887463569641,
+      "learning_rate": 0.0002,
+      "loss": 1.989,
+      "step": 700
+    },
+    {
+      "epoch": 1.6099773242630384,
+      "grad_norm": 0.6061141490936279,
+      "learning_rate": 0.0002,
+      "loss": 2.0419,
+      "step": 710
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 0.8291212320327759,
+      "learning_rate": 0.0002,
+      "loss": 2.0257,
+      "step": 720
+    },
+    {
+      "epoch": 1.655328798185941,
+      "grad_norm": 0.6672369241714478,
+      "learning_rate": 0.0002,
+      "loss": 2.0776,
+      "step": 730
+    },
+    {
+      "epoch": 1.6780045351473922,
+      "grad_norm": 0.6301238536834717,
+      "learning_rate": 0.0002,
+      "loss": 1.9834,
+      "step": 740
+    },
+    {
+      "epoch": 1.7006802721088436,
+      "grad_norm": 0.7217460870742798,
+      "learning_rate": 0.0002,
+      "loss": 2.0176,
+      "step": 750
+    },
+    {
+      "epoch": 1.7233560090702946,
+      "grad_norm": 0.6720302104949951,
+      "learning_rate": 0.0002,
+      "loss": 2.0793,
+      "step": 760
+    },
+    {
+      "epoch": 1.746031746031746,
+      "grad_norm": 0.8228253722190857,
+      "learning_rate": 0.0002,
+      "loss": 2.0417,
+      "step": 770
+    },
+    {
+      "epoch": 1.7687074829931972,
+      "grad_norm": 0.6359929442405701,
+      "learning_rate": 0.0002,
+      "loss": 2.0198,
+      "step": 780
+    },
+    {
+      "epoch": 1.7913832199546484,
+      "grad_norm": 0.6594814658164978,
+      "learning_rate": 0.0002,
+      "loss": 1.99,
+      "step": 790
+    },
+    {
+      "epoch": 1.8140589569160999,
+      "grad_norm": 0.6135239601135254,
+      "learning_rate": 0.0002,
+      "loss": 2.0752,
+      "step": 800
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 0.6736067533493042,
+      "learning_rate": 0.0002,
+      "loss": 2.0397,
+      "step": 810
+    },
+    {
+      "epoch": 1.8594104308390023,
+      "grad_norm": 0.7712678909301758,
+      "learning_rate": 0.0002,
+      "loss": 2.0545,
+      "step": 820
+    },
+    {
+      "epoch": 1.8820861678004537,
+      "grad_norm": 0.6950375437736511,
+      "learning_rate": 0.0002,
+      "loss": 1.9644,
+      "step": 830
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5913259387016296,
+      "learning_rate": 0.0002,
+      "loss": 1.9272,
+      "step": 840
+    },
+    {
+      "epoch": 1.927437641723356,
+      "grad_norm": 0.626690149307251,
+      "learning_rate": 0.0002,
+      "loss": 2.0068,
+      "step": 850
+    },
+    {
+      "epoch": 1.9501133786848073,
+      "grad_norm": 0.6986100673675537,
+      "learning_rate": 0.0002,
+      "loss": 2.0443,
+      "step": 860
+    },
+    {
+      "epoch": 1.9727891156462585,
+      "grad_norm": 0.6573156118392944,
+      "learning_rate": 0.0002,
+      "loss": 2.0585,
+      "step": 870
+    },
+    {
+      "epoch": 1.99546485260771,
+      "grad_norm": 0.7589707374572754,
+      "learning_rate": 0.0002,
+      "loss": 2.0454,
+      "step": 880
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.073181629180908,
+      "eval_runtime": 14.6998,
+      "eval_samples_per_second": 34.49,
+      "eval_steps_per_second": 4.354,
+      "step": 882
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3528,
+  "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": 1.0779800767561728e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/special_tokens_map.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.model b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer_config.json b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "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
+    }
+  },
+  "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-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_args.bin b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d9e930daf0ed2140cdc7c48443022c7b29e9e9f6
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bd01b61e91780113d7cfd71d7d1da24fc61dfbff2f52441a1dd4793ec50fda
+size 5560
diff --git a/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_log.jsonl b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..92e431da1ef72a374134be4d522cd891eac20519
--- /dev/null
+++ b/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 1.0, "step": 441, "epoch_duration": 241.79891800880432, "total_accumulated_duration": 241.79891800880432, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 5628.7490234375}, "avg_memory_reserved": {"GPU_0": 6182.0}, "peak_memory_reserved": {"GPU_0": 6182.0}, "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}]}
+{"epoch": 2.0, "step": 882, "epoch_duration": 214.3979184627533, "total_accumulated_duration": 456.1968364715576, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-441", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}]}
+{"epoch": 3.0, "step": 1323, "epoch_duration": 213.44339728355408, "total_accumulated_duration": 669.6402337551117, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-882", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}]}
+{"epoch": 4.0, "step": 1764, "epoch_duration": 227.5474293231964, "total_accumulated_duration": 897.1876630783081, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}, {"eval_loss": 2.061427116394043, "eval_runtime": 19.6731, "eval_samples_per_second": 25.771, "eval_steps_per_second": 3.253, "epoch": 3.0, "step": 1323}, {"loss": 1.8798, "grad_norm": 0.625606119632721, "learning_rate": 0.0002, "epoch": 3.015873015873016, "step": 1330}, {"loss": 1.8229, "grad_norm": 0.6679854989051819, "learning_rate": 0.0002, "epoch": 3.0385487528344672, "step": 1340}, {"loss": 1.8593, "grad_norm": 0.6756007671356201, "learning_rate": 0.0002, "epoch": 3.061224489795918, "step": 1350}, {"loss": 1.8126, "grad_norm": 0.8019710779190063, "learning_rate": 0.0002, "epoch": 3.0839002267573696, "step": 1360}, {"loss": 1.8593, "grad_norm": 0.9087499976158142, "learning_rate": 0.0002, "epoch": 3.106575963718821, "step": 1370}, {"loss": 1.8951, "grad_norm": 0.7627860903739929, "learning_rate": 0.0002, "epoch": 3.129251700680272, "step": 1380}, {"loss": 1.8308, "grad_norm": 0.761993944644928, "learning_rate": 0.0002, "epoch": 3.1519274376417235, "step": 1390}, {"loss": 1.9513, "grad_norm": 0.6993107795715332, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 1400}, {"loss": 1.8857, "grad_norm": 0.6398035287857056, "learning_rate": 0.0002, "epoch": 3.197278911564626, "step": 1410}, {"loss": 1.9041, "grad_norm": 0.8602580428123474, "learning_rate": 0.0002, "epoch": 3.2199546485260773, "step": 1420}, {"loss": 1.9629, "grad_norm": 0.7799932360649109, "learning_rate": 0.0002, "epoch": 3.2426303854875282, "step": 1430}, {"loss": 1.8352, "grad_norm": 0.6598666906356812, "learning_rate": 0.0002, "epoch": 3.2653061224489797, "step": 1440}, {"loss": 1.8772, "grad_norm": 0.6513776779174805, "learning_rate": 0.0002, "epoch": 3.287981859410431, "step": 1450}, {"loss": 1.8568, "grad_norm": 0.7690088748931885, "learning_rate": 0.0002, "epoch": 3.310657596371882, "step": 1460}, {"loss": 1.8535, "grad_norm": 1.0070042610168457, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1470}, {"loss": 1.8971, "grad_norm": 0.8606905341148376, "learning_rate": 0.0002, "epoch": 3.3560090702947845, "step": 1480}, {"loss": 1.8386, "grad_norm": 0.7596806287765503, "learning_rate": 0.0002, "epoch": 3.378684807256236, "step": 1490}, {"loss": 1.9278, "grad_norm": 0.6961908936500549, "learning_rate": 0.0002, "epoch": 3.4013605442176873, "step": 1500}, {"loss": 1.8908, "grad_norm": 0.8368703126907349, "learning_rate": 0.0002, "epoch": 3.4240362811791383, "step": 1510}, {"loss": 1.924, "grad_norm": 0.6655045747756958, "learning_rate": 0.0002, "epoch": 3.4467120181405897, "step": 1520}, {"loss": 1.8273, "grad_norm": 0.7557174563407898, "learning_rate": 0.0002, "epoch": 3.4693877551020407, "step": 1530}, {"loss": 1.8435, "grad_norm": 0.6713934540748596, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 1540}, {"loss": 1.9016, "grad_norm": 0.7006916999816895, "learning_rate": 0.0002, "epoch": 3.5147392290249435, "step": 1550}, {"loss": 1.9208, "grad_norm": 0.6728846430778503, "learning_rate": 0.0002, "epoch": 3.5374149659863945, "step": 1560}, {"loss": 1.8769, "grad_norm": 0.7479315400123596, "learning_rate": 0.0002, "epoch": 3.560090702947846, "step": 1570}, {"loss": 1.8558, "grad_norm": 0.7596876621246338, "learning_rate": 0.0002, "epoch": 3.582766439909297, "step": 1580}, {"loss": 1.876, "grad_norm": 0.8948362469673157, "learning_rate": 0.0002, "epoch": 3.6054421768707483, "step": 1590}, {"loss": 1.8552, "grad_norm": 0.6952481269836426, "learning_rate": 0.0002, "epoch": 3.6281179138321997, "step": 1600}, {"loss": 1.933, "grad_norm": 0.6494414210319519, "learning_rate": 0.0002, "epoch": 3.6507936507936507, "step": 1610}, {"loss": 1.8516, "grad_norm": 0.7438889145851135, "learning_rate": 0.0002, "epoch": 3.673469387755102, "step": 1620}, {"loss": 1.9441, "grad_norm": 0.7470757961273193, "learning_rate": 0.0002, "epoch": 3.696145124716553, "step": 1630}, {"loss": 1.8685, "grad_norm": 0.8727335333824158, "learning_rate": 0.0002, "epoch": 3.7188208616780045, "step": 1640}, {"loss": 1.8831, "grad_norm": 0.7064452767372131, "learning_rate": 0.0002, "epoch": 3.741496598639456, "step": 1650}, {"loss": 1.9743, "grad_norm": 0.6958122849464417, "learning_rate": 0.0002, "epoch": 3.764172335600907, "step": 1660}, {"loss": 1.927, "grad_norm": 0.6751590371131897, "learning_rate": 0.0002, "epoch": 3.7868480725623583, "step": 1670}, {"loss": 1.8868, "grad_norm": 0.7274497151374817, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 1680}, {"loss": 1.942, "grad_norm": 0.7311708331108093, "learning_rate": 0.0002, "epoch": 3.8321995464852607, "step": 1690}, {"loss": 1.8134, "grad_norm": 0.6703173518180847, "learning_rate": 0.0002, "epoch": 3.854875283446712, "step": 1700}, {"loss": 1.8687, "grad_norm": 0.6263390779495239, "learning_rate": 0.0002, "epoch": 3.877551020408163, "step": 1710}, {"loss": 1.9206, "grad_norm": 0.744022011756897, "learning_rate": 0.0002, "epoch": 3.9002267573696145, "step": 1720}, {"loss": 1.8961, "grad_norm": 0.7496722340583801, "learning_rate": 0.0002, "epoch": 3.9229024943310655, "step": 1730}, {"loss": 1.9212, "grad_norm": 0.6877635717391968, "learning_rate": 0.0002, "epoch": 3.945578231292517, "step": 1740}, {"loss": 1.8504, "grad_norm": 0.711262047290802, "learning_rate": 0.0002, "epoch": 3.9682539682539684, "step": 1750}, {"loss": 1.9315, "grad_norm": 0.7452033758163452, "learning_rate": 0.0002, "epoch": 3.9909297052154193, "step": 1760}]}
+{"epoch": 5.0, "step": 2205, "epoch_duration": 244.9496099948883, "total_accumulated_duration": 1142.1372730731964, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}, {"eval_loss": 2.061427116394043, "eval_runtime": 19.6731, "eval_samples_per_second": 25.771, "eval_steps_per_second": 3.253, "epoch": 3.0, "step": 1323}, {"loss": 1.8798, "grad_norm": 0.625606119632721, "learning_rate": 0.0002, "epoch": 3.015873015873016, "step": 1330}, {"loss": 1.8229, "grad_norm": 0.6679854989051819, "learning_rate": 0.0002, "epoch": 3.0385487528344672, "step": 1340}, {"loss": 1.8593, "grad_norm": 0.6756007671356201, "learning_rate": 0.0002, "epoch": 3.061224489795918, "step": 1350}, {"loss": 1.8126, "grad_norm": 0.8019710779190063, "learning_rate": 0.0002, "epoch": 3.0839002267573696, "step": 1360}, {"loss": 1.8593, "grad_norm": 0.9087499976158142, "learning_rate": 0.0002, "epoch": 3.106575963718821, "step": 1370}, {"loss": 1.8951, "grad_norm": 0.7627860903739929, "learning_rate": 0.0002, "epoch": 3.129251700680272, "step": 1380}, {"loss": 1.8308, "grad_norm": 0.761993944644928, "learning_rate": 0.0002, "epoch": 3.1519274376417235, "step": 1390}, {"loss": 1.9513, "grad_norm": 0.6993107795715332, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 1400}, {"loss": 1.8857, "grad_norm": 0.6398035287857056, "learning_rate": 0.0002, "epoch": 3.197278911564626, "step": 1410}, {"loss": 1.9041, "grad_norm": 0.8602580428123474, "learning_rate": 0.0002, "epoch": 3.2199546485260773, "step": 1420}, {"loss": 1.9629, "grad_norm": 0.7799932360649109, "learning_rate": 0.0002, "epoch": 3.2426303854875282, "step": 1430}, {"loss": 1.8352, "grad_norm": 0.6598666906356812, "learning_rate": 0.0002, "epoch": 3.2653061224489797, "step": 1440}, {"loss": 1.8772, "grad_norm": 0.6513776779174805, "learning_rate": 0.0002, "epoch": 3.287981859410431, "step": 1450}, {"loss": 1.8568, "grad_norm": 0.7690088748931885, "learning_rate": 0.0002, "epoch": 3.310657596371882, "step": 1460}, {"loss": 1.8535, "grad_norm": 1.0070042610168457, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1470}, {"loss": 1.8971, "grad_norm": 0.8606905341148376, "learning_rate": 0.0002, "epoch": 3.3560090702947845, "step": 1480}, {"loss": 1.8386, "grad_norm": 0.7596806287765503, "learning_rate": 0.0002, "epoch": 3.378684807256236, "step": 1490}, {"loss": 1.9278, "grad_norm": 0.6961908936500549, "learning_rate": 0.0002, "epoch": 3.4013605442176873, "step": 1500}, {"loss": 1.8908, "grad_norm": 0.8368703126907349, "learning_rate": 0.0002, "epoch": 3.4240362811791383, "step": 1510}, {"loss": 1.924, "grad_norm": 0.6655045747756958, "learning_rate": 0.0002, "epoch": 3.4467120181405897, "step": 1520}, {"loss": 1.8273, "grad_norm": 0.7557174563407898, "learning_rate": 0.0002, "epoch": 3.4693877551020407, "step": 1530}, {"loss": 1.8435, "grad_norm": 0.6713934540748596, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 1540}, {"loss": 1.9016, "grad_norm": 0.7006916999816895, "learning_rate": 0.0002, "epoch": 3.5147392290249435, "step": 1550}, {"loss": 1.9208, "grad_norm": 0.6728846430778503, "learning_rate": 0.0002, "epoch": 3.5374149659863945, "step": 1560}, {"loss": 1.8769, "grad_norm": 0.7479315400123596, "learning_rate": 0.0002, "epoch": 3.560090702947846, "step": 1570}, {"loss": 1.8558, "grad_norm": 0.7596876621246338, "learning_rate": 0.0002, "epoch": 3.582766439909297, "step": 1580}, {"loss": 1.876, "grad_norm": 0.8948362469673157, "learning_rate": 0.0002, "epoch": 3.6054421768707483, "step": 1590}, {"loss": 1.8552, "grad_norm": 0.6952481269836426, "learning_rate": 0.0002, "epoch": 3.6281179138321997, "step": 1600}, {"loss": 1.933, "grad_norm": 0.6494414210319519, "learning_rate": 0.0002, "epoch": 3.6507936507936507, "step": 1610}, {"loss": 1.8516, "grad_norm": 0.7438889145851135, "learning_rate": 0.0002, "epoch": 3.673469387755102, "step": 1620}, {"loss": 1.9441, "grad_norm": 0.7470757961273193, "learning_rate": 0.0002, "epoch": 3.696145124716553, "step": 1630}, {"loss": 1.8685, "grad_norm": 0.8727335333824158, "learning_rate": 0.0002, "epoch": 3.7188208616780045, "step": 1640}, {"loss": 1.8831, "grad_norm": 0.7064452767372131, "learning_rate": 0.0002, "epoch": 3.741496598639456, "step": 1650}, {"loss": 1.9743, "grad_norm": 0.6958122849464417, "learning_rate": 0.0002, "epoch": 3.764172335600907, "step": 1660}, {"loss": 1.927, "grad_norm": 0.6751590371131897, "learning_rate": 0.0002, "epoch": 3.7868480725623583, "step": 1670}, {"loss": 1.8868, "grad_norm": 0.7274497151374817, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 1680}, {"loss": 1.942, "grad_norm": 0.7311708331108093, "learning_rate": 0.0002, "epoch": 3.8321995464852607, "step": 1690}, {"loss": 1.8134, "grad_norm": 0.6703173518180847, "learning_rate": 0.0002, "epoch": 3.854875283446712, "step": 1700}, {"loss": 1.8687, "grad_norm": 0.6263390779495239, "learning_rate": 0.0002, "epoch": 3.877551020408163, "step": 1710}, {"loss": 1.9206, "grad_norm": 0.744022011756897, "learning_rate": 0.0002, "epoch": 3.9002267573696145, "step": 1720}, {"loss": 1.8961, "grad_norm": 0.7496722340583801, "learning_rate": 0.0002, "epoch": 3.9229024943310655, "step": 1730}, {"loss": 1.9212, "grad_norm": 0.6877635717391968, "learning_rate": 0.0002, "epoch": 3.945578231292517, "step": 1740}, {"loss": 1.8504, "grad_norm": 0.711262047290802, "learning_rate": 0.0002, "epoch": 3.9682539682539684, "step": 1750}, {"loss": 1.9315, "grad_norm": 0.7452033758163452, "learning_rate": 0.0002, "epoch": 3.9909297052154193, "step": 1760}, {"eval_loss": 2.065720319747925, "eval_runtime": 14.7103, "eval_samples_per_second": 34.466, "eval_steps_per_second": 4.351, "epoch": 4.0, "step": 1764}, {"loss": 1.8729, "grad_norm": 0.6913991570472717, "learning_rate": 0.0002, "epoch": 4.01360544217687, "step": 1770}, {"loss": 1.7617, "grad_norm": 0.7323354482650757, "learning_rate": 0.0002, "epoch": 4.036281179138322, "step": 1780}, {"loss": 1.7486, "grad_norm": 0.7691134214401245, "learning_rate": 0.0002, "epoch": 4.058956916099773, "step": 1790}, {"loss": 1.8102, "grad_norm": 0.7871596813201904, "learning_rate": 0.0002, "epoch": 4.081632653061225, "step": 1800}, {"loss": 1.7455, "grad_norm": 0.7079440355300903, "learning_rate": 0.0002, "epoch": 4.104308390022676, "step": 1810}, {"loss": 1.7693, "grad_norm": 0.7468329668045044, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 1820}, {"loss": 1.772, "grad_norm": 0.8017606139183044, "learning_rate": 0.0002, "epoch": 4.149659863945578, "step": 1830}, {"loss": 1.81, "grad_norm": 0.8591130375862122, "learning_rate": 0.0002, "epoch": 4.172335600907029, "step": 1840}, {"loss": 1.7356, "grad_norm": 0.7866499423980713, "learning_rate": 0.0002, "epoch": 4.195011337868481, "step": 1850}, {"loss": 1.8342, "grad_norm": 0.8457089066505432, "learning_rate": 0.0002, "epoch": 4.217687074829932, "step": 1860}, {"loss": 1.8744, "grad_norm": 0.8171738386154175, "learning_rate": 0.0002, "epoch": 4.240362811791384, "step": 1870}, {"loss": 1.7786, "grad_norm": 0.7377166748046875, "learning_rate": 0.0002, "epoch": 4.263038548752834, "step": 1880}, {"loss": 1.787, "grad_norm": 0.7925628423690796, "learning_rate": 0.0002, "epoch": 4.285714285714286, "step": 1890}, {"loss": 1.8634, "grad_norm": 1.0107476711273193, "learning_rate": 0.0002, "epoch": 4.308390022675737, "step": 1900}, {"loss": 1.8193, "grad_norm": 0.7791350483894348, "learning_rate": 0.0002, "epoch": 4.331065759637188, "step": 1910}, {"loss": 1.8431, "grad_norm": 0.7173832654953003, "learning_rate": 0.0002, "epoch": 4.35374149659864, "step": 1920}, {"loss": 1.83, "grad_norm": 0.8341949582099915, "learning_rate": 0.0002, "epoch": 4.37641723356009, "step": 1930}, {"loss": 1.7791, "grad_norm": 0.7944530844688416, "learning_rate": 0.0002, "epoch": 4.399092970521542, "step": 1940}, {"loss": 1.8161, "grad_norm": 0.7409125566482544, "learning_rate": 0.0002, "epoch": 4.421768707482993, "step": 1950}, {"loss": 1.7994, "grad_norm": 0.9393547773361206, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 1960}, {"loss": 1.8132, "grad_norm": 0.8280852437019348, "learning_rate": 0.0002, "epoch": 4.467120181405896, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.7249669432640076, "learning_rate": 0.0002, "epoch": 4.489795918367347, "step": 1980}, {"loss": 1.7846, "grad_norm": 0.7199458479881287, "learning_rate": 0.0002, "epoch": 4.512471655328798, "step": 1990}, {"loss": 1.7851, "grad_norm": 0.8454286456108093, "learning_rate": 0.0002, "epoch": 4.535147392290249, "step": 2000}, {"loss": 1.8673, "grad_norm": 0.7305117249488831, "learning_rate": 0.0002, "epoch": 4.557823129251701, "step": 2010}, {"loss": 1.8234, "grad_norm": 0.7819753885269165, "learning_rate": 0.0002, "epoch": 4.580498866213152, "step": 2020}, {"loss": 1.7972, "grad_norm": 0.8010004162788391, "learning_rate": 0.0002, "epoch": 4.603174603174603, "step": 2030}, {"loss": 1.7921, "grad_norm": 0.7850886583328247, "learning_rate": 0.0002, "epoch": 4.625850340136054, "step": 2040}, {"loss": 1.9024, "grad_norm": 0.7433428168296814, "learning_rate": 0.0002, "epoch": 4.648526077097506, "step": 2050}, {"loss": 1.8155, "grad_norm": 0.7518399357795715, "learning_rate": 0.0002, "epoch": 4.671201814058957, "step": 2060}, {"loss": 1.7924, "grad_norm": 0.7282178997993469, "learning_rate": 0.0002, "epoch": 4.6938775510204085, "step": 2070}, {"loss": 1.8144, "grad_norm": 0.6950743794441223, "learning_rate": 0.0002, "epoch": 4.71655328798186, "step": 2080}, {"loss": 1.7757, "grad_norm": 0.761830747127533, "learning_rate": 0.0002, "epoch": 4.73922902494331, "step": 2090}, {"loss": 1.8254, "grad_norm": 0.8068088293075562, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 2100}, {"loss": 1.8092, "grad_norm": 0.7580569386482239, "learning_rate": 0.0002, "epoch": 4.784580498866213, "step": 2110}, {"loss": 1.8144, "grad_norm": 0.9428760409355164, "learning_rate": 0.0002, "epoch": 4.807256235827665, "step": 2120}, {"loss": 1.829, "grad_norm": 0.7057452201843262, "learning_rate": 0.0002, "epoch": 4.829931972789115, "step": 2130}, {"loss": 1.7603, "grad_norm": 0.763630211353302, "learning_rate": 0.0002, "epoch": 4.852607709750567, "step": 2140}, {"loss": 1.8348, "grad_norm": 0.7298727035522461, "learning_rate": 0.0002, "epoch": 4.875283446712018, "step": 2150}, {"loss": 1.8652, "grad_norm": 0.7376292943954468, "learning_rate": 0.0002, "epoch": 4.8979591836734695, "step": 2160}, {"loss": 1.742, "grad_norm": 0.8528493642807007, "learning_rate": 0.0002, "epoch": 4.920634920634921, "step": 2170}, {"loss": 1.7706, "grad_norm": 0.7652511596679688, "learning_rate": 0.0002, "epoch": 4.943310657596372, "step": 2180}, {"loss": 1.7994, "grad_norm": 0.7602494955062866, "learning_rate": 0.0002, "epoch": 4.965986394557823, "step": 2190}, {"loss": 1.8278, "grad_norm": 0.758914053440094, "learning_rate": 0.0002, "epoch": 4.988662131519274, "step": 2200}]}
+{"epoch": 6.0, "step": 2646, "epoch_duration": 200.4551763534546, "total_accumulated_duration": 1342.592449426651, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}, {"eval_loss": 2.061427116394043, "eval_runtime": 19.6731, "eval_samples_per_second": 25.771, "eval_steps_per_second": 3.253, "epoch": 3.0, "step": 1323}, {"loss": 1.8798, "grad_norm": 0.625606119632721, "learning_rate": 0.0002, "epoch": 3.015873015873016, "step": 1330}, {"loss": 1.8229, "grad_norm": 0.6679854989051819, "learning_rate": 0.0002, "epoch": 3.0385487528344672, "step": 1340}, {"loss": 1.8593, "grad_norm": 0.6756007671356201, "learning_rate": 0.0002, "epoch": 3.061224489795918, "step": 1350}, {"loss": 1.8126, "grad_norm": 0.8019710779190063, "learning_rate": 0.0002, "epoch": 3.0839002267573696, "step": 1360}, {"loss": 1.8593, "grad_norm": 0.9087499976158142, "learning_rate": 0.0002, "epoch": 3.106575963718821, "step": 1370}, {"loss": 1.8951, "grad_norm": 0.7627860903739929, "learning_rate": 0.0002, "epoch": 3.129251700680272, "step": 1380}, {"loss": 1.8308, "grad_norm": 0.761993944644928, "learning_rate": 0.0002, "epoch": 3.1519274376417235, "step": 1390}, {"loss": 1.9513, "grad_norm": 0.6993107795715332, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 1400}, {"loss": 1.8857, "grad_norm": 0.6398035287857056, "learning_rate": 0.0002, "epoch": 3.197278911564626, "step": 1410}, {"loss": 1.9041, "grad_norm": 0.8602580428123474, "learning_rate": 0.0002, "epoch": 3.2199546485260773, "step": 1420}, {"loss": 1.9629, "grad_norm": 0.7799932360649109, "learning_rate": 0.0002, "epoch": 3.2426303854875282, "step": 1430}, {"loss": 1.8352, "grad_norm": 0.6598666906356812, "learning_rate": 0.0002, "epoch": 3.2653061224489797, "step": 1440}, {"loss": 1.8772, "grad_norm": 0.6513776779174805, "learning_rate": 0.0002, "epoch": 3.287981859410431, "step": 1450}, {"loss": 1.8568, "grad_norm": 0.7690088748931885, "learning_rate": 0.0002, "epoch": 3.310657596371882, "step": 1460}, {"loss": 1.8535, "grad_norm": 1.0070042610168457, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1470}, {"loss": 1.8971, "grad_norm": 0.8606905341148376, "learning_rate": 0.0002, "epoch": 3.3560090702947845, "step": 1480}, {"loss": 1.8386, "grad_norm": 0.7596806287765503, "learning_rate": 0.0002, "epoch": 3.378684807256236, "step": 1490}, {"loss": 1.9278, "grad_norm": 0.6961908936500549, "learning_rate": 0.0002, "epoch": 3.4013605442176873, "step": 1500}, {"loss": 1.8908, "grad_norm": 0.8368703126907349, "learning_rate": 0.0002, "epoch": 3.4240362811791383, "step": 1510}, {"loss": 1.924, "grad_norm": 0.6655045747756958, "learning_rate": 0.0002, "epoch": 3.4467120181405897, "step": 1520}, {"loss": 1.8273, "grad_norm": 0.7557174563407898, "learning_rate": 0.0002, "epoch": 3.4693877551020407, "step": 1530}, {"loss": 1.8435, "grad_norm": 0.6713934540748596, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 1540}, {"loss": 1.9016, "grad_norm": 0.7006916999816895, "learning_rate": 0.0002, "epoch": 3.5147392290249435, "step": 1550}, {"loss": 1.9208, "grad_norm": 0.6728846430778503, "learning_rate": 0.0002, "epoch": 3.5374149659863945, "step": 1560}, {"loss": 1.8769, "grad_norm": 0.7479315400123596, "learning_rate": 0.0002, "epoch": 3.560090702947846, "step": 1570}, {"loss": 1.8558, "grad_norm": 0.7596876621246338, "learning_rate": 0.0002, "epoch": 3.582766439909297, "step": 1580}, {"loss": 1.876, "grad_norm": 0.8948362469673157, "learning_rate": 0.0002, "epoch": 3.6054421768707483, "step": 1590}, {"loss": 1.8552, "grad_norm": 0.6952481269836426, "learning_rate": 0.0002, "epoch": 3.6281179138321997, "step": 1600}, {"loss": 1.933, "grad_norm": 0.6494414210319519, "learning_rate": 0.0002, "epoch": 3.6507936507936507, "step": 1610}, {"loss": 1.8516, "grad_norm": 0.7438889145851135, "learning_rate": 0.0002, "epoch": 3.673469387755102, "step": 1620}, {"loss": 1.9441, "grad_norm": 0.7470757961273193, "learning_rate": 0.0002, "epoch": 3.696145124716553, "step": 1630}, {"loss": 1.8685, "grad_norm": 0.8727335333824158, "learning_rate": 0.0002, "epoch": 3.7188208616780045, "step": 1640}, {"loss": 1.8831, "grad_norm": 0.7064452767372131, "learning_rate": 0.0002, "epoch": 3.741496598639456, "step": 1650}, {"loss": 1.9743, "grad_norm": 0.6958122849464417, "learning_rate": 0.0002, "epoch": 3.764172335600907, "step": 1660}, {"loss": 1.927, "grad_norm": 0.6751590371131897, "learning_rate": 0.0002, "epoch": 3.7868480725623583, "step": 1670}, {"loss": 1.8868, "grad_norm": 0.7274497151374817, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 1680}, {"loss": 1.942, "grad_norm": 0.7311708331108093, "learning_rate": 0.0002, "epoch": 3.8321995464852607, "step": 1690}, {"loss": 1.8134, "grad_norm": 0.6703173518180847, "learning_rate": 0.0002, "epoch": 3.854875283446712, "step": 1700}, {"loss": 1.8687, "grad_norm": 0.6263390779495239, "learning_rate": 0.0002, "epoch": 3.877551020408163, "step": 1710}, {"loss": 1.9206, "grad_norm": 0.744022011756897, "learning_rate": 0.0002, "epoch": 3.9002267573696145, "step": 1720}, {"loss": 1.8961, "grad_norm": 0.7496722340583801, "learning_rate": 0.0002, "epoch": 3.9229024943310655, "step": 1730}, {"loss": 1.9212, "grad_norm": 0.6877635717391968, "learning_rate": 0.0002, "epoch": 3.945578231292517, "step": 1740}, {"loss": 1.8504, "grad_norm": 0.711262047290802, "learning_rate": 0.0002, "epoch": 3.9682539682539684, "step": 1750}, {"loss": 1.9315, "grad_norm": 0.7452033758163452, "learning_rate": 0.0002, "epoch": 3.9909297052154193, "step": 1760}, {"eval_loss": 2.065720319747925, "eval_runtime": 14.7103, "eval_samples_per_second": 34.466, "eval_steps_per_second": 4.351, "epoch": 4.0, "step": 1764}, {"loss": 1.8729, "grad_norm": 0.6913991570472717, "learning_rate": 0.0002, "epoch": 4.01360544217687, "step": 1770}, {"loss": 1.7617, "grad_norm": 0.7323354482650757, "learning_rate": 0.0002, "epoch": 4.036281179138322, "step": 1780}, {"loss": 1.7486, "grad_norm": 0.7691134214401245, "learning_rate": 0.0002, "epoch": 4.058956916099773, "step": 1790}, {"loss": 1.8102, "grad_norm": 0.7871596813201904, "learning_rate": 0.0002, "epoch": 4.081632653061225, "step": 1800}, {"loss": 1.7455, "grad_norm": 0.7079440355300903, "learning_rate": 0.0002, "epoch": 4.104308390022676, "step": 1810}, {"loss": 1.7693, "grad_norm": 0.7468329668045044, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 1820}, {"loss": 1.772, "grad_norm": 0.8017606139183044, "learning_rate": 0.0002, "epoch": 4.149659863945578, "step": 1830}, {"loss": 1.81, "grad_norm": 0.8591130375862122, "learning_rate": 0.0002, "epoch": 4.172335600907029, "step": 1840}, {"loss": 1.7356, "grad_norm": 0.7866499423980713, "learning_rate": 0.0002, "epoch": 4.195011337868481, "step": 1850}, {"loss": 1.8342, "grad_norm": 0.8457089066505432, "learning_rate": 0.0002, "epoch": 4.217687074829932, "step": 1860}, {"loss": 1.8744, "grad_norm": 0.8171738386154175, "learning_rate": 0.0002, "epoch": 4.240362811791384, "step": 1870}, {"loss": 1.7786, "grad_norm": 0.7377166748046875, "learning_rate": 0.0002, "epoch": 4.263038548752834, "step": 1880}, {"loss": 1.787, "grad_norm": 0.7925628423690796, "learning_rate": 0.0002, "epoch": 4.285714285714286, "step": 1890}, {"loss": 1.8634, "grad_norm": 1.0107476711273193, "learning_rate": 0.0002, "epoch": 4.308390022675737, "step": 1900}, {"loss": 1.8193, "grad_norm": 0.7791350483894348, "learning_rate": 0.0002, "epoch": 4.331065759637188, "step": 1910}, {"loss": 1.8431, "grad_norm": 0.7173832654953003, "learning_rate": 0.0002, "epoch": 4.35374149659864, "step": 1920}, {"loss": 1.83, "grad_norm": 0.8341949582099915, "learning_rate": 0.0002, "epoch": 4.37641723356009, "step": 1930}, {"loss": 1.7791, "grad_norm": 0.7944530844688416, "learning_rate": 0.0002, "epoch": 4.399092970521542, "step": 1940}, {"loss": 1.8161, "grad_norm": 0.7409125566482544, "learning_rate": 0.0002, "epoch": 4.421768707482993, "step": 1950}, {"loss": 1.7994, "grad_norm": 0.9393547773361206, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 1960}, {"loss": 1.8132, "grad_norm": 0.8280852437019348, "learning_rate": 0.0002, "epoch": 4.467120181405896, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.7249669432640076, "learning_rate": 0.0002, "epoch": 4.489795918367347, "step": 1980}, {"loss": 1.7846, "grad_norm": 0.7199458479881287, "learning_rate": 0.0002, "epoch": 4.512471655328798, "step": 1990}, {"loss": 1.7851, "grad_norm": 0.8454286456108093, "learning_rate": 0.0002, "epoch": 4.535147392290249, "step": 2000}, {"loss": 1.8673, "grad_norm": 0.7305117249488831, "learning_rate": 0.0002, "epoch": 4.557823129251701, "step": 2010}, {"loss": 1.8234, "grad_norm": 0.7819753885269165, "learning_rate": 0.0002, "epoch": 4.580498866213152, "step": 2020}, {"loss": 1.7972, "grad_norm": 0.8010004162788391, "learning_rate": 0.0002, "epoch": 4.603174603174603, "step": 2030}, {"loss": 1.7921, "grad_norm": 0.7850886583328247, "learning_rate": 0.0002, "epoch": 4.625850340136054, "step": 2040}, {"loss": 1.9024, "grad_norm": 0.7433428168296814, "learning_rate": 0.0002, "epoch": 4.648526077097506, "step": 2050}, {"loss": 1.8155, "grad_norm": 0.7518399357795715, "learning_rate": 0.0002, "epoch": 4.671201814058957, "step": 2060}, {"loss": 1.7924, "grad_norm": 0.7282178997993469, "learning_rate": 0.0002, "epoch": 4.6938775510204085, "step": 2070}, {"loss": 1.8144, "grad_norm": 0.6950743794441223, "learning_rate": 0.0002, "epoch": 4.71655328798186, "step": 2080}, {"loss": 1.7757, "grad_norm": 0.761830747127533, "learning_rate": 0.0002, "epoch": 4.73922902494331, "step": 2090}, {"loss": 1.8254, "grad_norm": 0.8068088293075562, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 2100}, {"loss": 1.8092, "grad_norm": 0.7580569386482239, "learning_rate": 0.0002, "epoch": 4.784580498866213, "step": 2110}, {"loss": 1.8144, "grad_norm": 0.9428760409355164, "learning_rate": 0.0002, "epoch": 4.807256235827665, "step": 2120}, {"loss": 1.829, "grad_norm": 0.7057452201843262, "learning_rate": 0.0002, "epoch": 4.829931972789115, "step": 2130}, {"loss": 1.7603, "grad_norm": 0.763630211353302, "learning_rate": 0.0002, "epoch": 4.852607709750567, "step": 2140}, {"loss": 1.8348, "grad_norm": 0.7298727035522461, "learning_rate": 0.0002, "epoch": 4.875283446712018, "step": 2150}, {"loss": 1.8652, "grad_norm": 0.7376292943954468, "learning_rate": 0.0002, "epoch": 4.8979591836734695, "step": 2160}, {"loss": 1.742, "grad_norm": 0.8528493642807007, "learning_rate": 0.0002, "epoch": 4.920634920634921, "step": 2170}, {"loss": 1.7706, "grad_norm": 0.7652511596679688, "learning_rate": 0.0002, "epoch": 4.943310657596372, "step": 2180}, {"loss": 1.7994, "grad_norm": 0.7602494955062866, "learning_rate": 0.0002, "epoch": 4.965986394557823, "step": 2190}, {"loss": 1.8278, "grad_norm": 0.758914053440094, "learning_rate": 0.0002, "epoch": 4.988662131519274, "step": 2200}, {"eval_loss": 2.083296298980713, "eval_runtime": 20.2201, "eval_samples_per_second": 25.074, "eval_steps_per_second": 3.165, "epoch": 5.0, "step": 2205}, {"loss": 1.7003, "grad_norm": 0.6990088224411011, "learning_rate": 0.0002, "epoch": 5.011337868480726, "step": 2210}, {"loss": 1.6451, "grad_norm": 0.7627242803573608, "learning_rate": 0.0002, "epoch": 5.034013605442177, "step": 2220}, {"loss": 1.7713, "grad_norm": 0.8485700488090515, "learning_rate": 0.0002, "epoch": 5.0566893424036286, "step": 2230}, {"loss": 1.7143, "grad_norm": 0.8308766484260559, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 2240}, {"loss": 1.7318, "grad_norm": 0.825551450252533, "learning_rate": 0.0002, "epoch": 5.1020408163265305, "step": 2250}, {"loss": 1.7005, "grad_norm": 0.8571329116821289, "learning_rate": 0.0002, "epoch": 5.124716553287982, "step": 2260}, {"loss": 1.7304, "grad_norm": 0.7965090274810791, "learning_rate": 0.0002, "epoch": 5.147392290249433, "step": 2270}, {"loss": 1.713, "grad_norm": 0.7926788330078125, "learning_rate": 0.0002, "epoch": 5.170068027210885, "step": 2280}, {"loss": 1.7778, "grad_norm": 0.8400393128395081, "learning_rate": 0.0002, "epoch": 5.192743764172335, "step": 2290}, {"loss": 1.7163, "grad_norm": 0.9096018075942993, "learning_rate": 0.0002, "epoch": 5.215419501133787, "step": 2300}, {"loss": 1.7158, "grad_norm": 0.8156160712242126, "learning_rate": 0.0002, "epoch": 5.238095238095238, "step": 2310}, {"loss": 1.6941, "grad_norm": 0.8418492674827576, "learning_rate": 0.0002, "epoch": 5.26077097505669, "step": 2320}, {"loss": 1.7604, "grad_norm": 0.7994900345802307, "learning_rate": 0.0002, "epoch": 5.283446712018141, "step": 2330}, {"loss": 1.6869, "grad_norm": 0.9292947053909302, "learning_rate": 0.0002, "epoch": 5.3061224489795915, "step": 2340}, {"loss": 1.8213, "grad_norm": 0.8896278142929077, "learning_rate": 0.0002, "epoch": 5.328798185941043, "step": 2350}, {"loss": 1.6681, "grad_norm": 0.9424751400947571, "learning_rate": 0.0002, "epoch": 5.351473922902494, "step": 2360}, {"loss": 1.7032, "grad_norm": 0.8420370817184448, "learning_rate": 0.0002, "epoch": 5.374149659863946, "step": 2370}, {"loss": 1.665, "grad_norm": 0.8460527658462524, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 2380}, {"loss": 1.6443, "grad_norm": 1.0171664953231812, "learning_rate": 0.0002, "epoch": 5.419501133786848, "step": 2390}, {"loss": 1.744, "grad_norm": 0.9017615914344788, "learning_rate": 0.0002, "epoch": 5.442176870748299, "step": 2400}, {"loss": 1.7046, "grad_norm": 0.8650022745132446, "learning_rate": 0.0002, "epoch": 5.464852607709751, "step": 2410}, {"loss": 1.722, "grad_norm": 0.8252440094947815, "learning_rate": 0.0002, "epoch": 5.487528344671202, "step": 2420}, {"loss": 1.7473, "grad_norm": 0.9326912760734558, "learning_rate": 0.0002, "epoch": 5.510204081632653, "step": 2430}, {"loss": 1.7738, "grad_norm": 0.8913490772247314, "learning_rate": 0.0002, "epoch": 5.532879818594104, "step": 2440}, {"loss": 1.7197, "grad_norm": 0.8588173985481262, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 2450}, {"loss": 1.679, "grad_norm": 0.8768706321716309, "learning_rate": 0.0002, "epoch": 5.578231292517007, "step": 2460}, {"loss": 1.7005, "grad_norm": 0.9485137462615967, "learning_rate": 0.0002, "epoch": 5.600907029478458, "step": 2470}, {"loss": 1.6494, "grad_norm": 0.8462285399436951, "learning_rate": 0.0002, "epoch": 5.62358276643991, "step": 2480}, {"loss": 1.7199, "grad_norm": 0.8390767574310303, "learning_rate": 0.0002, "epoch": 5.646258503401361, "step": 2490}, {"loss": 1.7819, "grad_norm": 0.9576658010482788, "learning_rate": 0.0002, "epoch": 5.668934240362812, "step": 2500}, {"loss": 1.6684, "grad_norm": 0.9528797268867493, "learning_rate": 0.0002, "epoch": 5.691609977324263, "step": 2510}, {"loss": 1.6769, "grad_norm": 1.19835364818573, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2520}, {"loss": 1.6454, "grad_norm": 0.8822922110557556, "learning_rate": 0.0002, "epoch": 5.736961451247166, "step": 2530}, {"loss": 1.7769, "grad_norm": 0.9812421202659607, "learning_rate": 0.0002, "epoch": 5.759637188208616, "step": 2540}, {"loss": 1.6781, "grad_norm": 0.9178062677383423, "learning_rate": 0.0002, "epoch": 5.782312925170068, "step": 2550}, {"loss": 1.685, "grad_norm": 1.0217735767364502, "learning_rate": 0.0002, "epoch": 5.804988662131519, "step": 2560}, {"loss": 1.722, "grad_norm": 0.8624402284622192, "learning_rate": 0.0002, "epoch": 5.827664399092971, "step": 2570}, {"loss": 1.6976, "grad_norm": 0.8644461035728455, "learning_rate": 0.0002, "epoch": 5.850340136054422, "step": 2580}, {"loss": 1.7498, "grad_norm": 0.8437298536300659, "learning_rate": 0.0002, "epoch": 5.8730158730158735, "step": 2590}, {"loss": 1.7577, "grad_norm": 0.7952260971069336, "learning_rate": 0.0002, "epoch": 5.895691609977324, "step": 2600}, {"loss": 1.743, "grad_norm": 0.9374001622200012, "learning_rate": 0.0002, "epoch": 5.918367346938775, "step": 2610}, {"loss": 1.7689, "grad_norm": 0.8157470226287842, "learning_rate": 0.0002, "epoch": 5.941043083900227, "step": 2620}, {"loss": 1.6757, "grad_norm": 0.8078457713127136, "learning_rate": 0.0002, "epoch": 5.963718820861678, "step": 2630}, {"loss": 1.7059, "grad_norm": 0.8926942348480225, "learning_rate": 0.0002, "epoch": 5.986394557823129, "step": 2640}]}
+{"epoch": 7.0, "step": 3087, "epoch_duration": 185.4541997909546, "total_accumulated_duration": 1528.0466492176056, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}, {"eval_loss": 2.061427116394043, "eval_runtime": 19.6731, "eval_samples_per_second": 25.771, "eval_steps_per_second": 3.253, "epoch": 3.0, "step": 1323}, {"loss": 1.8798, "grad_norm": 0.625606119632721, "learning_rate": 0.0002, "epoch": 3.015873015873016, "step": 1330}, {"loss": 1.8229, "grad_norm": 0.6679854989051819, "learning_rate": 0.0002, "epoch": 3.0385487528344672, "step": 1340}, {"loss": 1.8593, "grad_norm": 0.6756007671356201, "learning_rate": 0.0002, "epoch": 3.061224489795918, "step": 1350}, {"loss": 1.8126, "grad_norm": 0.8019710779190063, "learning_rate": 0.0002, "epoch": 3.0839002267573696, "step": 1360}, {"loss": 1.8593, "grad_norm": 0.9087499976158142, "learning_rate": 0.0002, "epoch": 3.106575963718821, "step": 1370}, {"loss": 1.8951, "grad_norm": 0.7627860903739929, "learning_rate": 0.0002, "epoch": 3.129251700680272, "step": 1380}, {"loss": 1.8308, "grad_norm": 0.761993944644928, "learning_rate": 0.0002, "epoch": 3.1519274376417235, "step": 1390}, {"loss": 1.9513, "grad_norm": 0.6993107795715332, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 1400}, {"loss": 1.8857, "grad_norm": 0.6398035287857056, "learning_rate": 0.0002, "epoch": 3.197278911564626, "step": 1410}, {"loss": 1.9041, "grad_norm": 0.8602580428123474, "learning_rate": 0.0002, "epoch": 3.2199546485260773, "step": 1420}, {"loss": 1.9629, "grad_norm": 0.7799932360649109, "learning_rate": 0.0002, "epoch": 3.2426303854875282, "step": 1430}, {"loss": 1.8352, "grad_norm": 0.6598666906356812, "learning_rate": 0.0002, "epoch": 3.2653061224489797, "step": 1440}, {"loss": 1.8772, "grad_norm": 0.6513776779174805, "learning_rate": 0.0002, "epoch": 3.287981859410431, "step": 1450}, {"loss": 1.8568, "grad_norm": 0.7690088748931885, "learning_rate": 0.0002, "epoch": 3.310657596371882, "step": 1460}, {"loss": 1.8535, "grad_norm": 1.0070042610168457, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1470}, {"loss": 1.8971, "grad_norm": 0.8606905341148376, "learning_rate": 0.0002, "epoch": 3.3560090702947845, "step": 1480}, {"loss": 1.8386, "grad_norm": 0.7596806287765503, "learning_rate": 0.0002, "epoch": 3.378684807256236, "step": 1490}, {"loss": 1.9278, "grad_norm": 0.6961908936500549, "learning_rate": 0.0002, "epoch": 3.4013605442176873, "step": 1500}, {"loss": 1.8908, "grad_norm": 0.8368703126907349, "learning_rate": 0.0002, "epoch": 3.4240362811791383, "step": 1510}, {"loss": 1.924, "grad_norm": 0.6655045747756958, "learning_rate": 0.0002, "epoch": 3.4467120181405897, "step": 1520}, {"loss": 1.8273, "grad_norm": 0.7557174563407898, "learning_rate": 0.0002, "epoch": 3.4693877551020407, "step": 1530}, {"loss": 1.8435, "grad_norm": 0.6713934540748596, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 1540}, {"loss": 1.9016, "grad_norm": 0.7006916999816895, "learning_rate": 0.0002, "epoch": 3.5147392290249435, "step": 1550}, {"loss": 1.9208, "grad_norm": 0.6728846430778503, "learning_rate": 0.0002, "epoch": 3.5374149659863945, "step": 1560}, {"loss": 1.8769, "grad_norm": 0.7479315400123596, "learning_rate": 0.0002, "epoch": 3.560090702947846, "step": 1570}, {"loss": 1.8558, "grad_norm": 0.7596876621246338, "learning_rate": 0.0002, "epoch": 3.582766439909297, "step": 1580}, {"loss": 1.876, "grad_norm": 0.8948362469673157, "learning_rate": 0.0002, "epoch": 3.6054421768707483, "step": 1590}, {"loss": 1.8552, "grad_norm": 0.6952481269836426, "learning_rate": 0.0002, "epoch": 3.6281179138321997, "step": 1600}, {"loss": 1.933, "grad_norm": 0.6494414210319519, "learning_rate": 0.0002, "epoch": 3.6507936507936507, "step": 1610}, {"loss": 1.8516, "grad_norm": 0.7438889145851135, "learning_rate": 0.0002, "epoch": 3.673469387755102, "step": 1620}, {"loss": 1.9441, "grad_norm": 0.7470757961273193, "learning_rate": 0.0002, "epoch": 3.696145124716553, "step": 1630}, {"loss": 1.8685, "grad_norm": 0.8727335333824158, "learning_rate": 0.0002, "epoch": 3.7188208616780045, "step": 1640}, {"loss": 1.8831, "grad_norm": 0.7064452767372131, "learning_rate": 0.0002, "epoch": 3.741496598639456, "step": 1650}, {"loss": 1.9743, "grad_norm": 0.6958122849464417, "learning_rate": 0.0002, "epoch": 3.764172335600907, "step": 1660}, {"loss": 1.927, "grad_norm": 0.6751590371131897, "learning_rate": 0.0002, "epoch": 3.7868480725623583, "step": 1670}, {"loss": 1.8868, "grad_norm": 0.7274497151374817, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 1680}, {"loss": 1.942, "grad_norm": 0.7311708331108093, "learning_rate": 0.0002, "epoch": 3.8321995464852607, "step": 1690}, {"loss": 1.8134, "grad_norm": 0.6703173518180847, "learning_rate": 0.0002, "epoch": 3.854875283446712, "step": 1700}, {"loss": 1.8687, "grad_norm": 0.6263390779495239, "learning_rate": 0.0002, "epoch": 3.877551020408163, "step": 1710}, {"loss": 1.9206, "grad_norm": 0.744022011756897, "learning_rate": 0.0002, "epoch": 3.9002267573696145, "step": 1720}, {"loss": 1.8961, "grad_norm": 0.7496722340583801, "learning_rate": 0.0002, "epoch": 3.9229024943310655, "step": 1730}, {"loss": 1.9212, "grad_norm": 0.6877635717391968, "learning_rate": 0.0002, "epoch": 3.945578231292517, "step": 1740}, {"loss": 1.8504, "grad_norm": 0.711262047290802, "learning_rate": 0.0002, "epoch": 3.9682539682539684, "step": 1750}, {"loss": 1.9315, "grad_norm": 0.7452033758163452, "learning_rate": 0.0002, "epoch": 3.9909297052154193, "step": 1760}, {"eval_loss": 2.065720319747925, "eval_runtime": 14.7103, "eval_samples_per_second": 34.466, "eval_steps_per_second": 4.351, "epoch": 4.0, "step": 1764}, {"loss": 1.8729, "grad_norm": 0.6913991570472717, "learning_rate": 0.0002, "epoch": 4.01360544217687, "step": 1770}, {"loss": 1.7617, "grad_norm": 0.7323354482650757, "learning_rate": 0.0002, "epoch": 4.036281179138322, "step": 1780}, {"loss": 1.7486, "grad_norm": 0.7691134214401245, "learning_rate": 0.0002, "epoch": 4.058956916099773, "step": 1790}, {"loss": 1.8102, "grad_norm": 0.7871596813201904, "learning_rate": 0.0002, "epoch": 4.081632653061225, "step": 1800}, {"loss": 1.7455, "grad_norm": 0.7079440355300903, "learning_rate": 0.0002, "epoch": 4.104308390022676, "step": 1810}, {"loss": 1.7693, "grad_norm": 0.7468329668045044, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 1820}, {"loss": 1.772, "grad_norm": 0.8017606139183044, "learning_rate": 0.0002, "epoch": 4.149659863945578, "step": 1830}, {"loss": 1.81, "grad_norm": 0.8591130375862122, "learning_rate": 0.0002, "epoch": 4.172335600907029, "step": 1840}, {"loss": 1.7356, "grad_norm": 0.7866499423980713, "learning_rate": 0.0002, "epoch": 4.195011337868481, "step": 1850}, {"loss": 1.8342, "grad_norm": 0.8457089066505432, "learning_rate": 0.0002, "epoch": 4.217687074829932, "step": 1860}, {"loss": 1.8744, "grad_norm": 0.8171738386154175, "learning_rate": 0.0002, "epoch": 4.240362811791384, "step": 1870}, {"loss": 1.7786, "grad_norm": 0.7377166748046875, "learning_rate": 0.0002, "epoch": 4.263038548752834, "step": 1880}, {"loss": 1.787, "grad_norm": 0.7925628423690796, "learning_rate": 0.0002, "epoch": 4.285714285714286, "step": 1890}, {"loss": 1.8634, "grad_norm": 1.0107476711273193, "learning_rate": 0.0002, "epoch": 4.308390022675737, "step": 1900}, {"loss": 1.8193, "grad_norm": 0.7791350483894348, "learning_rate": 0.0002, "epoch": 4.331065759637188, "step": 1910}, {"loss": 1.8431, "grad_norm": 0.7173832654953003, "learning_rate": 0.0002, "epoch": 4.35374149659864, "step": 1920}, {"loss": 1.83, "grad_norm": 0.8341949582099915, "learning_rate": 0.0002, "epoch": 4.37641723356009, "step": 1930}, {"loss": 1.7791, "grad_norm": 0.7944530844688416, "learning_rate": 0.0002, "epoch": 4.399092970521542, "step": 1940}, {"loss": 1.8161, "grad_norm": 0.7409125566482544, "learning_rate": 0.0002, "epoch": 4.421768707482993, "step": 1950}, {"loss": 1.7994, "grad_norm": 0.9393547773361206, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 1960}, {"loss": 1.8132, "grad_norm": 0.8280852437019348, "learning_rate": 0.0002, "epoch": 4.467120181405896, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.7249669432640076, "learning_rate": 0.0002, "epoch": 4.489795918367347, "step": 1980}, {"loss": 1.7846, "grad_norm": 0.7199458479881287, "learning_rate": 0.0002, "epoch": 4.512471655328798, "step": 1990}, {"loss": 1.7851, "grad_norm": 0.8454286456108093, "learning_rate": 0.0002, "epoch": 4.535147392290249, "step": 2000}, {"loss": 1.8673, "grad_norm": 0.7305117249488831, "learning_rate": 0.0002, "epoch": 4.557823129251701, "step": 2010}, {"loss": 1.8234, "grad_norm": 0.7819753885269165, "learning_rate": 0.0002, "epoch": 4.580498866213152, "step": 2020}, {"loss": 1.7972, "grad_norm": 0.8010004162788391, "learning_rate": 0.0002, "epoch": 4.603174603174603, "step": 2030}, {"loss": 1.7921, "grad_norm": 0.7850886583328247, "learning_rate": 0.0002, "epoch": 4.625850340136054, "step": 2040}, {"loss": 1.9024, "grad_norm": 0.7433428168296814, "learning_rate": 0.0002, "epoch": 4.648526077097506, "step": 2050}, {"loss": 1.8155, "grad_norm": 0.7518399357795715, "learning_rate": 0.0002, "epoch": 4.671201814058957, "step": 2060}, {"loss": 1.7924, "grad_norm": 0.7282178997993469, "learning_rate": 0.0002, "epoch": 4.6938775510204085, "step": 2070}, {"loss": 1.8144, "grad_norm": 0.6950743794441223, "learning_rate": 0.0002, "epoch": 4.71655328798186, "step": 2080}, {"loss": 1.7757, "grad_norm": 0.761830747127533, "learning_rate": 0.0002, "epoch": 4.73922902494331, "step": 2090}, {"loss": 1.8254, "grad_norm": 0.8068088293075562, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 2100}, {"loss": 1.8092, "grad_norm": 0.7580569386482239, "learning_rate": 0.0002, "epoch": 4.784580498866213, "step": 2110}, {"loss": 1.8144, "grad_norm": 0.9428760409355164, "learning_rate": 0.0002, "epoch": 4.807256235827665, "step": 2120}, {"loss": 1.829, "grad_norm": 0.7057452201843262, "learning_rate": 0.0002, "epoch": 4.829931972789115, "step": 2130}, {"loss": 1.7603, "grad_norm": 0.763630211353302, "learning_rate": 0.0002, "epoch": 4.852607709750567, "step": 2140}, {"loss": 1.8348, "grad_norm": 0.7298727035522461, "learning_rate": 0.0002, "epoch": 4.875283446712018, "step": 2150}, {"loss": 1.8652, "grad_norm": 0.7376292943954468, "learning_rate": 0.0002, "epoch": 4.8979591836734695, "step": 2160}, {"loss": 1.742, "grad_norm": 0.8528493642807007, "learning_rate": 0.0002, "epoch": 4.920634920634921, "step": 2170}, {"loss": 1.7706, "grad_norm": 0.7652511596679688, "learning_rate": 0.0002, "epoch": 4.943310657596372, "step": 2180}, {"loss": 1.7994, "grad_norm": 0.7602494955062866, "learning_rate": 0.0002, "epoch": 4.965986394557823, "step": 2190}, {"loss": 1.8278, "grad_norm": 0.758914053440094, "learning_rate": 0.0002, "epoch": 4.988662131519274, "step": 2200}, {"eval_loss": 2.083296298980713, "eval_runtime": 20.2201, "eval_samples_per_second": 25.074, "eval_steps_per_second": 3.165, "epoch": 5.0, "step": 2205}, {"loss": 1.7003, "grad_norm": 0.6990088224411011, "learning_rate": 0.0002, "epoch": 5.011337868480726, "step": 2210}, {"loss": 1.6451, "grad_norm": 0.7627242803573608, "learning_rate": 0.0002, "epoch": 5.034013605442177, "step": 2220}, {"loss": 1.7713, "grad_norm": 0.8485700488090515, "learning_rate": 0.0002, "epoch": 5.0566893424036286, "step": 2230}, {"loss": 1.7143, "grad_norm": 0.8308766484260559, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 2240}, {"loss": 1.7318, "grad_norm": 0.825551450252533, "learning_rate": 0.0002, "epoch": 5.1020408163265305, "step": 2250}, {"loss": 1.7005, "grad_norm": 0.8571329116821289, "learning_rate": 0.0002, "epoch": 5.124716553287982, "step": 2260}, {"loss": 1.7304, "grad_norm": 0.7965090274810791, "learning_rate": 0.0002, "epoch": 5.147392290249433, "step": 2270}, {"loss": 1.713, "grad_norm": 0.7926788330078125, "learning_rate": 0.0002, "epoch": 5.170068027210885, "step": 2280}, {"loss": 1.7778, "grad_norm": 0.8400393128395081, "learning_rate": 0.0002, "epoch": 5.192743764172335, "step": 2290}, {"loss": 1.7163, "grad_norm": 0.9096018075942993, "learning_rate": 0.0002, "epoch": 5.215419501133787, "step": 2300}, {"loss": 1.7158, "grad_norm": 0.8156160712242126, "learning_rate": 0.0002, "epoch": 5.238095238095238, "step": 2310}, {"loss": 1.6941, "grad_norm": 0.8418492674827576, "learning_rate": 0.0002, "epoch": 5.26077097505669, "step": 2320}, {"loss": 1.7604, "grad_norm": 0.7994900345802307, "learning_rate": 0.0002, "epoch": 5.283446712018141, "step": 2330}, {"loss": 1.6869, "grad_norm": 0.9292947053909302, "learning_rate": 0.0002, "epoch": 5.3061224489795915, "step": 2340}, {"loss": 1.8213, "grad_norm": 0.8896278142929077, "learning_rate": 0.0002, "epoch": 5.328798185941043, "step": 2350}, {"loss": 1.6681, "grad_norm": 0.9424751400947571, "learning_rate": 0.0002, "epoch": 5.351473922902494, "step": 2360}, {"loss": 1.7032, "grad_norm": 0.8420370817184448, "learning_rate": 0.0002, "epoch": 5.374149659863946, "step": 2370}, {"loss": 1.665, "grad_norm": 0.8460527658462524, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 2380}, {"loss": 1.6443, "grad_norm": 1.0171664953231812, "learning_rate": 0.0002, "epoch": 5.419501133786848, "step": 2390}, {"loss": 1.744, "grad_norm": 0.9017615914344788, "learning_rate": 0.0002, "epoch": 5.442176870748299, "step": 2400}, {"loss": 1.7046, "grad_norm": 0.8650022745132446, "learning_rate": 0.0002, "epoch": 5.464852607709751, "step": 2410}, {"loss": 1.722, "grad_norm": 0.8252440094947815, "learning_rate": 0.0002, "epoch": 5.487528344671202, "step": 2420}, {"loss": 1.7473, "grad_norm": 0.9326912760734558, "learning_rate": 0.0002, "epoch": 5.510204081632653, "step": 2430}, {"loss": 1.7738, "grad_norm": 0.8913490772247314, "learning_rate": 0.0002, "epoch": 5.532879818594104, "step": 2440}, {"loss": 1.7197, "grad_norm": 0.8588173985481262, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 2450}, {"loss": 1.679, "grad_norm": 0.8768706321716309, "learning_rate": 0.0002, "epoch": 5.578231292517007, "step": 2460}, {"loss": 1.7005, "grad_norm": 0.9485137462615967, "learning_rate": 0.0002, "epoch": 5.600907029478458, "step": 2470}, {"loss": 1.6494, "grad_norm": 0.8462285399436951, "learning_rate": 0.0002, "epoch": 5.62358276643991, "step": 2480}, {"loss": 1.7199, "grad_norm": 0.8390767574310303, "learning_rate": 0.0002, "epoch": 5.646258503401361, "step": 2490}, {"loss": 1.7819, "grad_norm": 0.9576658010482788, "learning_rate": 0.0002, "epoch": 5.668934240362812, "step": 2500}, {"loss": 1.6684, "grad_norm": 0.9528797268867493, "learning_rate": 0.0002, "epoch": 5.691609977324263, "step": 2510}, {"loss": 1.6769, "grad_norm": 1.19835364818573, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2520}, {"loss": 1.6454, "grad_norm": 0.8822922110557556, "learning_rate": 0.0002, "epoch": 5.736961451247166, "step": 2530}, {"loss": 1.7769, "grad_norm": 0.9812421202659607, "learning_rate": 0.0002, "epoch": 5.759637188208616, "step": 2540}, {"loss": 1.6781, "grad_norm": 0.9178062677383423, "learning_rate": 0.0002, "epoch": 5.782312925170068, "step": 2550}, {"loss": 1.685, "grad_norm": 1.0217735767364502, "learning_rate": 0.0002, "epoch": 5.804988662131519, "step": 2560}, {"loss": 1.722, "grad_norm": 0.8624402284622192, "learning_rate": 0.0002, "epoch": 5.827664399092971, "step": 2570}, {"loss": 1.6976, "grad_norm": 0.8644461035728455, "learning_rate": 0.0002, "epoch": 5.850340136054422, "step": 2580}, {"loss": 1.7498, "grad_norm": 0.8437298536300659, "learning_rate": 0.0002, "epoch": 5.8730158730158735, "step": 2590}, {"loss": 1.7577, "grad_norm": 0.7952260971069336, "learning_rate": 0.0002, "epoch": 5.895691609977324, "step": 2600}, {"loss": 1.743, "grad_norm": 0.9374001622200012, "learning_rate": 0.0002, "epoch": 5.918367346938775, "step": 2610}, {"loss": 1.7689, "grad_norm": 0.8157470226287842, "learning_rate": 0.0002, "epoch": 5.941043083900227, "step": 2620}, {"loss": 1.6757, "grad_norm": 0.8078457713127136, "learning_rate": 0.0002, "epoch": 5.963718820861678, "step": 2630}, {"loss": 1.7059, "grad_norm": 0.8926942348480225, "learning_rate": 0.0002, "epoch": 5.986394557823129, "step": 2640}, {"eval_loss": 2.117249011993408, "eval_runtime": 14.7114, "eval_samples_per_second": 34.463, "eval_steps_per_second": 4.35, "epoch": 6.0, "step": 2646}, {"loss": 1.6913, "grad_norm": 0.866627037525177, "learning_rate": 0.0002, "epoch": 6.00907029478458, "step": 2650}, {"loss": 1.5567, "grad_norm": 0.9636926054954529, "learning_rate": 0.0002, "epoch": 6.031746031746032, "step": 2660}, {"loss": 1.5617, "grad_norm": 0.9994971752166748, "learning_rate": 0.0002, "epoch": 6.054421768707483, "step": 2670}, {"loss": 1.5724, "grad_norm": 0.886381983757019, "learning_rate": 0.0002, "epoch": 6.0770975056689345, "step": 2680}, {"loss": 1.7039, "grad_norm": 0.8596521615982056, "learning_rate": 0.0002, "epoch": 6.099773242630386, "step": 2690}, {"loss": 1.6268, "grad_norm": 1.0034657716751099, "learning_rate": 0.0002, "epoch": 6.122448979591836, "step": 2700}, {"loss": 1.5562, "grad_norm": 0.8961828947067261, "learning_rate": 0.0002, "epoch": 6.145124716553288, "step": 2710}, {"loss": 1.5924, "grad_norm": 0.9269746541976929, "learning_rate": 0.0002, "epoch": 6.167800453514739, "step": 2720}, {"loss": 1.6335, "grad_norm": 0.9434096217155457, "learning_rate": 0.0002, "epoch": 6.190476190476191, "step": 2730}, {"loss": 1.6189, "grad_norm": 0.9009348750114441, "learning_rate": 0.0002, "epoch": 6.213151927437642, "step": 2740}, {"loss": 1.6322, "grad_norm": 1.0572991371154785, "learning_rate": 0.0002, "epoch": 6.235827664399093, "step": 2750}, {"loss": 1.6675, "grad_norm": 0.996005654335022, "learning_rate": 0.0002, "epoch": 6.258503401360544, "step": 2760}, {"loss": 1.5836, "grad_norm": 0.8857560753822327, "learning_rate": 0.0002, "epoch": 6.2811791383219955, "step": 2770}, {"loss": 1.6365, "grad_norm": 0.9528077840805054, "learning_rate": 0.0002, "epoch": 6.303854875283447, "step": 2780}, {"loss": 1.636, "grad_norm": 0.8799851536750793, "learning_rate": 0.0002, "epoch": 6.326530612244898, "step": 2790}, {"loss": 1.6227, "grad_norm": 0.9826605916023254, "learning_rate": 0.0002, "epoch": 6.349206349206349, "step": 2800}, {"loss": 1.5672, "grad_norm": 1.0445730686187744, "learning_rate": 0.0002, "epoch": 6.3718820861678, "step": 2810}, {"loss": 1.6476, "grad_norm": 1.0388351678848267, "learning_rate": 0.0002, "epoch": 6.394557823129252, "step": 2820}, {"loss": 1.5719, "grad_norm": 0.9709566831588745, "learning_rate": 0.0002, "epoch": 6.417233560090703, "step": 2830}, {"loss": 1.6654, "grad_norm": 0.9056107401847839, "learning_rate": 0.0002, "epoch": 6.4399092970521545, "step": 2840}, {"loss": 1.6482, "grad_norm": 1.0178978443145752, "learning_rate": 0.0002, "epoch": 6.462585034013605, "step": 2850}, {"loss": 1.5917, "grad_norm": 0.9525032639503479, "learning_rate": 0.0002, "epoch": 6.4852607709750565, "step": 2860}, {"loss": 1.6261, "grad_norm": 0.9033688306808472, "learning_rate": 0.0002, "epoch": 6.507936507936508, "step": 2870}, {"loss": 1.5718, "grad_norm": 0.9108299016952515, "learning_rate": 0.0002, "epoch": 6.530612244897959, "step": 2880}, {"loss": 1.6426, "grad_norm": 0.961175262928009, "learning_rate": 0.0002, "epoch": 6.553287981859411, "step": 2890}, {"loss": 1.5573, "grad_norm": 0.8874754309654236, "learning_rate": 0.0002, "epoch": 6.575963718820862, "step": 2900}, {"loss": 1.6003, "grad_norm": 0.9972973465919495, "learning_rate": 0.0002, "epoch": 6.598639455782313, "step": 2910}, {"loss": 1.5926, "grad_norm": 0.9361436367034912, "learning_rate": 0.0002, "epoch": 6.621315192743764, "step": 2920}, {"loss": 1.6858, "grad_norm": 0.9379439353942871, "learning_rate": 0.0002, "epoch": 6.6439909297052155, "step": 2930}, {"loss": 1.6558, "grad_norm": 0.9896080493927002, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 2940}, {"loss": 1.6045, "grad_norm": 1.0048781633377075, "learning_rate": 0.0002, "epoch": 6.6893424036281175, "step": 2950}, {"loss": 1.6269, "grad_norm": 0.9979572892189026, "learning_rate": 0.0002, "epoch": 6.712018140589569, "step": 2960}, {"loss": 1.6663, "grad_norm": 0.9461016654968262, "learning_rate": 0.0002, "epoch": 6.73469387755102, "step": 2970}, {"loss": 1.6948, "grad_norm": 0.9460693597793579, "learning_rate": 0.0002, "epoch": 6.757369614512472, "step": 2980}, {"loss": 1.5654, "grad_norm": 0.9581215977668762, "learning_rate": 0.0002, "epoch": 6.780045351473923, "step": 2990}, {"loss": 1.6814, "grad_norm": 0.8998590111732483, "learning_rate": 0.0002, "epoch": 6.802721088435375, "step": 3000}, {"loss": 1.6812, "grad_norm": 0.9226139187812805, "learning_rate": 0.0002, "epoch": 6.825396825396825, "step": 3010}, {"loss": 1.6407, "grad_norm": 0.9358452558517456, "learning_rate": 0.0002, "epoch": 6.8480725623582765, "step": 3020}, {"loss": 1.6654, "grad_norm": 1.1139743328094482, "learning_rate": 0.0002, "epoch": 6.870748299319728, "step": 3030}, {"loss": 1.6326, "grad_norm": 0.9886892437934875, "learning_rate": 0.0002, "epoch": 6.893424036281179, "step": 3040}, {"loss": 1.6329, "grad_norm": 0.885146975517273, "learning_rate": 0.0002, "epoch": 6.91609977324263, "step": 3050}, {"loss": 1.6473, "grad_norm": 0.9668041467666626, "learning_rate": 0.0002, "epoch": 6.938775510204081, "step": 3060}, {"loss": 1.5385, "grad_norm": 0.9532362222671509, "learning_rate": 0.0002, "epoch": 6.961451247165533, "step": 3070}, {"loss": 1.5744, "grad_norm": 1.0338832139968872, "learning_rate": 0.0002, "epoch": 6.984126984126984, "step": 3080}]}
+{"epoch": 8.0, "step": 3528, "epoch_duration": 198.99810981750488, "total_accumulated_duration": 1727.0447590351105, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 20170.0}, "peak_memory_reserved": {"GPU_0": 20170.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-1/checkpoint-1323", "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": 4.7982, "grad_norm": 1.8457136154174805, "learning_rate": 0.0002, "epoch": 0.022675736961451247, "step": 10}, {"loss": 3.71, "grad_norm": 1.0967960357666016, "learning_rate": 0.0002, "epoch": 0.045351473922902494, "step": 20}, {"loss": 3.3086, "grad_norm": 0.8257759213447571, "learning_rate": 0.0002, "epoch": 0.06802721088435375, "step": 30}, {"loss": 2.923, "grad_norm": 0.8248435258865356, "learning_rate": 0.0002, "epoch": 0.09070294784580499, "step": 40}, {"loss": 2.8263, "grad_norm": 1.1190264225006104, "learning_rate": 0.0002, "epoch": 0.11337868480725624, "step": 50}, {"loss": 2.6239, "grad_norm": 1.273951768875122, "learning_rate": 0.0002, "epoch": 0.1360544217687075, "step": 60}, {"loss": 2.5203, "grad_norm": 1.0004316568374634, "learning_rate": 0.0002, "epoch": 0.15873015873015872, "step": 70}, {"loss": 2.5055, "grad_norm": 0.8296012878417969, "learning_rate": 0.0002, "epoch": 0.18140589569160998, "step": 80}, {"loss": 2.5255, "grad_norm": 1.1563128232955933, "learning_rate": 0.0002, "epoch": 0.20408163265306123, "step": 90}, {"loss": 2.4731, "grad_norm": 0.7863451838493347, "learning_rate": 0.0002, "epoch": 0.22675736961451248, "step": 100}, {"loss": 2.3688, "grad_norm": 0.9167478680610657, "learning_rate": 0.0002, "epoch": 0.2494331065759637, "step": 110}, {"loss": 2.336, "grad_norm": 1.104406476020813, "learning_rate": 0.0002, "epoch": 0.272108843537415, "step": 120}, {"loss": 2.3063, "grad_norm": 1.109564185142517, "learning_rate": 0.0002, "epoch": 0.2947845804988662, "step": 130}, {"loss": 2.23, "grad_norm": 0.7416750192642212, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 140}, {"loss": 2.3346, "grad_norm": 1.1499258279800415, "learning_rate": 0.0002, "epoch": 0.3401360544217687, "step": 150}, {"loss": 2.2853, "grad_norm": 0.9961133599281311, "learning_rate": 0.0002, "epoch": 0.36281179138321995, "step": 160}, {"loss": 2.2731, "grad_norm": 0.9257232546806335, "learning_rate": 0.0002, "epoch": 0.3854875283446712, "step": 170}, {"loss": 2.1986, "grad_norm": 1.098513126373291, "learning_rate": 0.0002, "epoch": 0.40816326530612246, "step": 180}, {"loss": 2.1685, "grad_norm": 0.8935237526893616, "learning_rate": 0.0002, "epoch": 0.4308390022675737, "step": 190}, {"loss": 2.2077, "grad_norm": 0.9654302597045898, "learning_rate": 0.0002, "epoch": 0.45351473922902497, "step": 200}, {"loss": 2.2421, "grad_norm": 0.9273215532302856, "learning_rate": 0.0002, "epoch": 0.47619047619047616, "step": 210}, {"loss": 2.1973, "grad_norm": 0.7222742438316345, "learning_rate": 0.0002, "epoch": 0.4988662131519274, "step": 220}, {"loss": 2.0619, "grad_norm": 0.7825018763542175, "learning_rate": 0.0002, "epoch": 0.5215419501133787, "step": 230}, {"loss": 2.2247, "grad_norm": 0.7758685350418091, "learning_rate": 0.0002, "epoch": 0.54421768707483, "step": 240}, {"loss": 2.1518, "grad_norm": 0.916719913482666, "learning_rate": 0.0002, "epoch": 0.5668934240362812, "step": 250}, {"loss": 2.0614, "grad_norm": 0.696332573890686, "learning_rate": 0.0002, "epoch": 0.5895691609977324, "step": 260}, {"loss": 2.0763, "grad_norm": 0.7362682223320007, "learning_rate": 0.0002, "epoch": 0.6122448979591837, "step": 270}, {"loss": 2.223, "grad_norm": 0.8184517025947571, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 280}, {"loss": 2.1634, "grad_norm": 0.7944937348365784, "learning_rate": 0.0002, "epoch": 0.6575963718820862, "step": 290}, {"loss": 2.0922, "grad_norm": 0.7258933186531067, "learning_rate": 0.0002, "epoch": 0.6802721088435374, "step": 300}, {"loss": 2.1221, "grad_norm": 0.7037094831466675, "learning_rate": 0.0002, "epoch": 0.7029478458049887, "step": 310}, {"loss": 2.1132, "grad_norm": 0.7499093413352966, "learning_rate": 0.0002, "epoch": 0.7256235827664399, "step": 320}, {"loss": 2.2321, "grad_norm": 0.7725803256034851, "learning_rate": 0.0002, "epoch": 0.7482993197278912, "step": 330}, {"loss": 2.1611, "grad_norm": 0.6771686673164368, "learning_rate": 0.0002, "epoch": 0.7709750566893424, "step": 340}, {"loss": 2.0693, "grad_norm": 0.7557646036148071, "learning_rate": 0.0002, "epoch": 0.7936507936507936, "step": 350}, {"loss": 2.0629, "grad_norm": 0.7430537939071655, "learning_rate": 0.0002, "epoch": 0.8163265306122449, "step": 360}, {"loss": 2.0731, "grad_norm": 0.8602280616760254, "learning_rate": 0.0002, "epoch": 0.8390022675736961, "step": 370}, {"loss": 2.1836, "grad_norm": 0.753440260887146, "learning_rate": 0.0002, "epoch": 0.8616780045351474, "step": 380}, {"loss": 2.1186, "grad_norm": 0.7217596769332886, "learning_rate": 0.0002, "epoch": 0.8843537414965986, "step": 390}, {"loss": 2.0223, "grad_norm": 0.6116304397583008, "learning_rate": 0.0002, "epoch": 0.9070294784580499, "step": 400}, {"loss": 2.0491, "grad_norm": 0.6829515695571899, "learning_rate": 0.0002, "epoch": 0.9297052154195011, "step": 410}, {"loss": 2.1518, "grad_norm": 0.733259916305542, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 420}, {"loss": 2.076, "grad_norm": 0.6036455035209656, "learning_rate": 0.0002, "epoch": 0.9750566893424036, "step": 430}, {"loss": 2.1526, "grad_norm": 0.8478217720985413, "learning_rate": 0.0002, "epoch": 0.9977324263038548, "step": 440}, {"eval_loss": 2.114302158355713, "eval_runtime": 16.5108, "eval_samples_per_second": 30.707, "eval_steps_per_second": 3.876, "epoch": 1.0, "step": 441}, {"loss": 1.9967, "grad_norm": 0.6996896266937256, "learning_rate": 0.0002, "epoch": 1.0204081632653061, "step": 450}, {"loss": 2.0471, "grad_norm": 0.7616267800331116, "learning_rate": 0.0002, "epoch": 1.0430839002267573, "step": 460}, {"loss": 2.0674, "grad_norm": 0.5978519320487976, "learning_rate": 0.0002, "epoch": 1.0657596371882085, "step": 470}, {"loss": 2.0803, "grad_norm": 0.703987717628479, "learning_rate": 0.0002, "epoch": 1.08843537414966, "step": 480}, {"loss": 2.0215, "grad_norm": 0.6707539558410645, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 490}, {"loss": 2.045, "grad_norm": 0.6343862414360046, "learning_rate": 0.0002, "epoch": 1.1337868480725624, "step": 500}, {"loss": 1.9753, "grad_norm": 0.8836050033569336, "learning_rate": 0.0002, "epoch": 1.1564625850340136, "step": 510}, {"loss": 1.9537, "grad_norm": 0.6288692951202393, "learning_rate": 0.0002, "epoch": 1.179138321995465, "step": 520}, {"loss": 2.0072, "grad_norm": 0.6671382784843445, "learning_rate": 0.0002, "epoch": 1.2018140589569162, "step": 530}, {"loss": 2.0631, "grad_norm": 0.7260547876358032, "learning_rate": 0.0002, "epoch": 1.2244897959183674, "step": 540}, {"loss": 2.1395, "grad_norm": 0.6919140219688416, "learning_rate": 0.0002, "epoch": 1.2471655328798186, "step": 550}, {"loss": 2.1162, "grad_norm": 0.6895381212234497, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 560}, {"loss": 2.1023, "grad_norm": 0.7131213545799255, "learning_rate": 0.0002, "epoch": 1.2925170068027212, "step": 570}, {"loss": 2.0381, "grad_norm": 0.6087185144424438, "learning_rate": 0.0002, "epoch": 1.3151927437641724, "step": 580}, {"loss": 2.054, "grad_norm": 0.6506251096725464, "learning_rate": 0.0002, "epoch": 1.3378684807256236, "step": 590}, {"loss": 2.0669, "grad_norm": 0.6999365091323853, "learning_rate": 0.0002, "epoch": 1.3605442176870748, "step": 600}, {"loss": 2.0121, "grad_norm": 0.6305016875267029, "learning_rate": 0.0002, "epoch": 1.383219954648526, "step": 610}, {"loss": 1.9727, "grad_norm": 0.6413465142250061, "learning_rate": 0.0002, "epoch": 1.4058956916099774, "step": 620}, {"loss": 2.0676, "grad_norm": 0.7571105360984802, "learning_rate": 0.0002, "epoch": 1.4285714285714286, "step": 630}, {"loss": 2.0838, "grad_norm": 0.8087857365608215, "learning_rate": 0.0002, "epoch": 1.4512471655328798, "step": 640}, {"loss": 1.9932, "grad_norm": 0.6284025311470032, "learning_rate": 0.0002, "epoch": 1.473922902494331, "step": 650}, {"loss": 2.0456, "grad_norm": 0.6630339026451111, "learning_rate": 0.0002, "epoch": 1.4965986394557822, "step": 660}, {"loss": 2.0778, "grad_norm": 0.6777389645576477, "learning_rate": 0.0002, "epoch": 1.5192743764172336, "step": 670}, {"loss": 2.0437, "grad_norm": 0.6965354681015015, "learning_rate": 0.0002, "epoch": 1.5419501133786848, "step": 680}, {"loss": 2.056, "grad_norm": 0.7124104499816895, "learning_rate": 0.0002, "epoch": 1.564625850340136, "step": 690}, {"loss": 1.989, "grad_norm": 0.6652887463569641, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 700}, {"loss": 2.0419, "grad_norm": 0.6061141490936279, "learning_rate": 0.0002, "epoch": 1.6099773242630384, "step": 710}, {"loss": 2.0257, "grad_norm": 0.8291212320327759, "learning_rate": 0.0002, "epoch": 1.6326530612244898, "step": 720}, {"loss": 2.0776, "grad_norm": 0.6672369241714478, "learning_rate": 0.0002, "epoch": 1.655328798185941, "step": 730}, {"loss": 1.9834, "grad_norm": 0.6301238536834717, "learning_rate": 0.0002, "epoch": 1.6780045351473922, "step": 740}, {"loss": 2.0176, "grad_norm": 0.7217460870742798, "learning_rate": 0.0002, "epoch": 1.7006802721088436, "step": 750}, {"loss": 2.0793, "grad_norm": 0.6720302104949951, "learning_rate": 0.0002, "epoch": 1.7233560090702946, "step": 760}, {"loss": 2.0417, "grad_norm": 0.8228253722190857, "learning_rate": 0.0002, "epoch": 1.746031746031746, "step": 770}, {"loss": 2.0198, "grad_norm": 0.6359929442405701, "learning_rate": 0.0002, "epoch": 1.7687074829931972, "step": 780}, {"loss": 1.99, "grad_norm": 0.6594814658164978, "learning_rate": 0.0002, "epoch": 1.7913832199546484, "step": 790}, {"loss": 2.0752, "grad_norm": 0.6135239601135254, "learning_rate": 0.0002, "epoch": 1.8140589569160999, "step": 800}, {"loss": 2.0397, "grad_norm": 0.6736067533493042, "learning_rate": 0.0002, "epoch": 1.836734693877551, "step": 810}, {"loss": 2.0545, "grad_norm": 0.7712678909301758, "learning_rate": 0.0002, "epoch": 1.8594104308390023, "step": 820}, {"loss": 1.9644, "grad_norm": 0.6950375437736511, "learning_rate": 0.0002, "epoch": 1.8820861678004537, "step": 830}, {"loss": 1.9272, "grad_norm": 0.5913259387016296, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 840}, {"loss": 2.0068, "grad_norm": 0.626690149307251, "learning_rate": 0.0002, "epoch": 1.927437641723356, "step": 850}, {"loss": 2.0443, "grad_norm": 0.6986100673675537, "learning_rate": 0.0002, "epoch": 1.9501133786848073, "step": 860}, {"loss": 2.0585, "grad_norm": 0.6573156118392944, "learning_rate": 0.0002, "epoch": 1.9727891156462585, "step": 870}, {"loss": 2.0454, "grad_norm": 0.7589707374572754, "learning_rate": 0.0002, "epoch": 1.99546485260771, "step": 880}, {"eval_loss": 2.073181629180908, "eval_runtime": 14.6998, "eval_samples_per_second": 34.49, "eval_steps_per_second": 4.354, "epoch": 2.0, "step": 882}, {"loss": 2.015, "grad_norm": 0.7971214056015015, "learning_rate": 0.0002, "epoch": 2.018140589569161, "step": 890}, {"loss": 1.982, "grad_norm": 0.6144835948944092, "learning_rate": 0.0002, "epoch": 2.0408163265306123, "step": 900}, {"loss": 1.9391, "grad_norm": 0.7324559688568115, "learning_rate": 0.0002, "epoch": 2.0634920634920633, "step": 910}, {"loss": 1.9066, "grad_norm": 0.6465244293212891, "learning_rate": 0.0002, "epoch": 2.0861678004535147, "step": 920}, {"loss": 1.9893, "grad_norm": 0.7424528002738953, "learning_rate": 0.0002, "epoch": 2.108843537414966, "step": 930}, {"loss": 1.9193, "grad_norm": 0.6717734336853027, "learning_rate": 0.0002, "epoch": 2.131519274376417, "step": 940}, {"loss": 1.9569, "grad_norm": 0.7873768210411072, "learning_rate": 0.0002, "epoch": 2.1541950113378685, "step": 950}, {"loss": 1.9563, "grad_norm": 0.6752461194992065, "learning_rate": 0.0002, "epoch": 2.17687074829932, "step": 960}, {"loss": 2.0406, "grad_norm": 0.6919583082199097, "learning_rate": 0.0002, "epoch": 2.199546485260771, "step": 970}, {"loss": 1.9281, "grad_norm": 0.6355509161949158, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 980}, {"loss": 1.9953, "grad_norm": 0.770007312297821, "learning_rate": 0.0002, "epoch": 2.2448979591836733, "step": 990}, {"loss": 1.9589, "grad_norm": 0.6216670870780945, "learning_rate": 0.0002, "epoch": 2.2675736961451247, "step": 1000}, {"loss": 1.9247, "grad_norm": 0.6880573034286499, "learning_rate": 0.0002, "epoch": 2.290249433106576, "step": 1010}, {"loss": 1.9426, "grad_norm": 0.7346075177192688, "learning_rate": 0.0002, "epoch": 2.312925170068027, "step": 1020}, {"loss": 1.9461, "grad_norm": 0.6651889681816101, "learning_rate": 0.0002, "epoch": 2.3356009070294785, "step": 1030}, {"loss": 1.9147, "grad_norm": 0.7324103713035583, "learning_rate": 0.0002, "epoch": 2.35827664399093, "step": 1040}, {"loss": 1.9142, "grad_norm": 0.7186149954795837, "learning_rate": 0.0002, "epoch": 2.380952380952381, "step": 1050}, {"loss": 1.9501, "grad_norm": 0.6849178671836853, "learning_rate": 0.0002, "epoch": 2.4036281179138324, "step": 1060}, {"loss": 1.889, "grad_norm": 0.6259638667106628, "learning_rate": 0.0002, "epoch": 2.4263038548752833, "step": 1070}, {"loss": 1.9446, "grad_norm": 0.6821024417877197, "learning_rate": 0.0002, "epoch": 2.4489795918367347, "step": 1080}, {"loss": 1.9398, "grad_norm": 0.7019884586334229, "learning_rate": 0.0002, "epoch": 2.471655328798186, "step": 1090}, {"loss": 1.9325, "grad_norm": 0.6368714570999146, "learning_rate": 0.0002, "epoch": 2.494331065759637, "step": 1100}, {"loss": 1.9773, "grad_norm": 0.6608313918113708, "learning_rate": 0.0002, "epoch": 2.5170068027210886, "step": 1110}, {"loss": 1.9337, "grad_norm": 0.6855525970458984, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 1120}, {"loss": 1.9347, "grad_norm": 0.7279609441757202, "learning_rate": 0.0002, "epoch": 2.562358276643991, "step": 1130}, {"loss": 1.9508, "grad_norm": 0.6539294123649597, "learning_rate": 0.0002, "epoch": 2.5850340136054424, "step": 1140}, {"loss": 2.0517, "grad_norm": 0.7035511136054993, "learning_rate": 0.0002, "epoch": 2.6077097505668934, "step": 1150}, {"loss": 1.9955, "grad_norm": 0.7031794786453247, "learning_rate": 0.0002, "epoch": 2.630385487528345, "step": 1160}, {"loss": 1.9549, "grad_norm": 0.7587798833847046, "learning_rate": 0.0002, "epoch": 2.6530612244897958, "step": 1170}, {"loss": 2.0014, "grad_norm": 0.707339882850647, "learning_rate": 0.0002, "epoch": 2.675736961451247, "step": 1180}, {"loss": 1.955, "grad_norm": 0.6633496880531311, "learning_rate": 0.0002, "epoch": 2.6984126984126986, "step": 1190}, {"loss": 1.9434, "grad_norm": 0.9233061671257019, "learning_rate": 0.0002, "epoch": 2.7210884353741496, "step": 1200}, {"loss": 1.9146, "grad_norm": 0.6396011710166931, "learning_rate": 0.0002, "epoch": 2.743764172335601, "step": 1210}, {"loss": 1.9934, "grad_norm": 0.6500924229621887, "learning_rate": 0.0002, "epoch": 2.766439909297052, "step": 1220}, {"loss": 1.9197, "grad_norm": 0.6185927391052246, "learning_rate": 0.0002, "epoch": 2.7891156462585034, "step": 1230}, {"loss": 1.9809, "grad_norm": 0.6737818717956543, "learning_rate": 0.0002, "epoch": 2.811791383219955, "step": 1240}, {"loss": 1.9508, "grad_norm": 0.8148701786994934, "learning_rate": 0.0002, "epoch": 2.834467120181406, "step": 1250}, {"loss": 1.9848, "grad_norm": 0.7563779354095459, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1260}, {"loss": 1.9822, "grad_norm": 0.6679276823997498, "learning_rate": 0.0002, "epoch": 2.879818594104308, "step": 1270}, {"loss": 2.0609, "grad_norm": 0.7446274757385254, "learning_rate": 0.0002, "epoch": 2.9024943310657596, "step": 1280}, {"loss": 1.9251, "grad_norm": 0.6530991196632385, "learning_rate": 0.0002, "epoch": 2.925170068027211, "step": 1290}, {"loss": 1.9555, "grad_norm": 0.6491292119026184, "learning_rate": 0.0002, "epoch": 2.947845804988662, "step": 1300}, {"loss": 1.913, "grad_norm": 0.6711354851722717, "learning_rate": 0.0002, "epoch": 2.9705215419501134, "step": 1310}, {"loss": 2.0005, "grad_norm": 0.6327161192893982, "learning_rate": 0.0002, "epoch": 2.9931972789115644, "step": 1320}, {"eval_loss": 2.061427116394043, "eval_runtime": 19.6731, "eval_samples_per_second": 25.771, "eval_steps_per_second": 3.253, "epoch": 3.0, "step": 1323}, {"loss": 1.8798, "grad_norm": 0.625606119632721, "learning_rate": 0.0002, "epoch": 3.015873015873016, "step": 1330}, {"loss": 1.8229, "grad_norm": 0.6679854989051819, "learning_rate": 0.0002, "epoch": 3.0385487528344672, "step": 1340}, {"loss": 1.8593, "grad_norm": 0.6756007671356201, "learning_rate": 0.0002, "epoch": 3.061224489795918, "step": 1350}, {"loss": 1.8126, "grad_norm": 0.8019710779190063, "learning_rate": 0.0002, "epoch": 3.0839002267573696, "step": 1360}, {"loss": 1.8593, "grad_norm": 0.9087499976158142, "learning_rate": 0.0002, "epoch": 3.106575963718821, "step": 1370}, {"loss": 1.8951, "grad_norm": 0.7627860903739929, "learning_rate": 0.0002, "epoch": 3.129251700680272, "step": 1380}, {"loss": 1.8308, "grad_norm": 0.761993944644928, "learning_rate": 0.0002, "epoch": 3.1519274376417235, "step": 1390}, {"loss": 1.9513, "grad_norm": 0.6993107795715332, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 1400}, {"loss": 1.8857, "grad_norm": 0.6398035287857056, "learning_rate": 0.0002, "epoch": 3.197278911564626, "step": 1410}, {"loss": 1.9041, "grad_norm": 0.8602580428123474, "learning_rate": 0.0002, "epoch": 3.2199546485260773, "step": 1420}, {"loss": 1.9629, "grad_norm": 0.7799932360649109, "learning_rate": 0.0002, "epoch": 3.2426303854875282, "step": 1430}, {"loss": 1.8352, "grad_norm": 0.6598666906356812, "learning_rate": 0.0002, "epoch": 3.2653061224489797, "step": 1440}, {"loss": 1.8772, "grad_norm": 0.6513776779174805, "learning_rate": 0.0002, "epoch": 3.287981859410431, "step": 1450}, {"loss": 1.8568, "grad_norm": 0.7690088748931885, "learning_rate": 0.0002, "epoch": 3.310657596371882, "step": 1460}, {"loss": 1.8535, "grad_norm": 1.0070042610168457, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 1470}, {"loss": 1.8971, "grad_norm": 0.8606905341148376, "learning_rate": 0.0002, "epoch": 3.3560090702947845, "step": 1480}, {"loss": 1.8386, "grad_norm": 0.7596806287765503, "learning_rate": 0.0002, "epoch": 3.378684807256236, "step": 1490}, {"loss": 1.9278, "grad_norm": 0.6961908936500549, "learning_rate": 0.0002, "epoch": 3.4013605442176873, "step": 1500}, {"loss": 1.8908, "grad_norm": 0.8368703126907349, "learning_rate": 0.0002, "epoch": 3.4240362811791383, "step": 1510}, {"loss": 1.924, "grad_norm": 0.6655045747756958, "learning_rate": 0.0002, "epoch": 3.4467120181405897, "step": 1520}, {"loss": 1.8273, "grad_norm": 0.7557174563407898, "learning_rate": 0.0002, "epoch": 3.4693877551020407, "step": 1530}, {"loss": 1.8435, "grad_norm": 0.6713934540748596, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 1540}, {"loss": 1.9016, "grad_norm": 0.7006916999816895, "learning_rate": 0.0002, "epoch": 3.5147392290249435, "step": 1550}, {"loss": 1.9208, "grad_norm": 0.6728846430778503, "learning_rate": 0.0002, "epoch": 3.5374149659863945, "step": 1560}, {"loss": 1.8769, "grad_norm": 0.7479315400123596, "learning_rate": 0.0002, "epoch": 3.560090702947846, "step": 1570}, {"loss": 1.8558, "grad_norm": 0.7596876621246338, "learning_rate": 0.0002, "epoch": 3.582766439909297, "step": 1580}, {"loss": 1.876, "grad_norm": 0.8948362469673157, "learning_rate": 0.0002, "epoch": 3.6054421768707483, "step": 1590}, {"loss": 1.8552, "grad_norm": 0.6952481269836426, "learning_rate": 0.0002, "epoch": 3.6281179138321997, "step": 1600}, {"loss": 1.933, "grad_norm": 0.6494414210319519, "learning_rate": 0.0002, "epoch": 3.6507936507936507, "step": 1610}, {"loss": 1.8516, "grad_norm": 0.7438889145851135, "learning_rate": 0.0002, "epoch": 3.673469387755102, "step": 1620}, {"loss": 1.9441, "grad_norm": 0.7470757961273193, "learning_rate": 0.0002, "epoch": 3.696145124716553, "step": 1630}, {"loss": 1.8685, "grad_norm": 0.8727335333824158, "learning_rate": 0.0002, "epoch": 3.7188208616780045, "step": 1640}, {"loss": 1.8831, "grad_norm": 0.7064452767372131, "learning_rate": 0.0002, "epoch": 3.741496598639456, "step": 1650}, {"loss": 1.9743, "grad_norm": 0.6958122849464417, "learning_rate": 0.0002, "epoch": 3.764172335600907, "step": 1660}, {"loss": 1.927, "grad_norm": 0.6751590371131897, "learning_rate": 0.0002, "epoch": 3.7868480725623583, "step": 1670}, {"loss": 1.8868, "grad_norm": 0.7274497151374817, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 1680}, {"loss": 1.942, "grad_norm": 0.7311708331108093, "learning_rate": 0.0002, "epoch": 3.8321995464852607, "step": 1690}, {"loss": 1.8134, "grad_norm": 0.6703173518180847, "learning_rate": 0.0002, "epoch": 3.854875283446712, "step": 1700}, {"loss": 1.8687, "grad_norm": 0.6263390779495239, "learning_rate": 0.0002, "epoch": 3.877551020408163, "step": 1710}, {"loss": 1.9206, "grad_norm": 0.744022011756897, "learning_rate": 0.0002, "epoch": 3.9002267573696145, "step": 1720}, {"loss": 1.8961, "grad_norm": 0.7496722340583801, "learning_rate": 0.0002, "epoch": 3.9229024943310655, "step": 1730}, {"loss": 1.9212, "grad_norm": 0.6877635717391968, "learning_rate": 0.0002, "epoch": 3.945578231292517, "step": 1740}, {"loss": 1.8504, "grad_norm": 0.711262047290802, "learning_rate": 0.0002, "epoch": 3.9682539682539684, "step": 1750}, {"loss": 1.9315, "grad_norm": 0.7452033758163452, "learning_rate": 0.0002, "epoch": 3.9909297052154193, "step": 1760}, {"eval_loss": 2.065720319747925, "eval_runtime": 14.7103, "eval_samples_per_second": 34.466, "eval_steps_per_second": 4.351, "epoch": 4.0, "step": 1764}, {"loss": 1.8729, "grad_norm": 0.6913991570472717, "learning_rate": 0.0002, "epoch": 4.01360544217687, "step": 1770}, {"loss": 1.7617, "grad_norm": 0.7323354482650757, "learning_rate": 0.0002, "epoch": 4.036281179138322, "step": 1780}, {"loss": 1.7486, "grad_norm": 0.7691134214401245, "learning_rate": 0.0002, "epoch": 4.058956916099773, "step": 1790}, {"loss": 1.8102, "grad_norm": 0.7871596813201904, "learning_rate": 0.0002, "epoch": 4.081632653061225, "step": 1800}, {"loss": 1.7455, "grad_norm": 0.7079440355300903, "learning_rate": 0.0002, "epoch": 4.104308390022676, "step": 1810}, {"loss": 1.7693, "grad_norm": 0.7468329668045044, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 1820}, {"loss": 1.772, "grad_norm": 0.8017606139183044, "learning_rate": 0.0002, "epoch": 4.149659863945578, "step": 1830}, {"loss": 1.81, "grad_norm": 0.8591130375862122, "learning_rate": 0.0002, "epoch": 4.172335600907029, "step": 1840}, {"loss": 1.7356, "grad_norm": 0.7866499423980713, "learning_rate": 0.0002, "epoch": 4.195011337868481, "step": 1850}, {"loss": 1.8342, "grad_norm": 0.8457089066505432, "learning_rate": 0.0002, "epoch": 4.217687074829932, "step": 1860}, {"loss": 1.8744, "grad_norm": 0.8171738386154175, "learning_rate": 0.0002, "epoch": 4.240362811791384, "step": 1870}, {"loss": 1.7786, "grad_norm": 0.7377166748046875, "learning_rate": 0.0002, "epoch": 4.263038548752834, "step": 1880}, {"loss": 1.787, "grad_norm": 0.7925628423690796, "learning_rate": 0.0002, "epoch": 4.285714285714286, "step": 1890}, {"loss": 1.8634, "grad_norm": 1.0107476711273193, "learning_rate": 0.0002, "epoch": 4.308390022675737, "step": 1900}, {"loss": 1.8193, "grad_norm": 0.7791350483894348, "learning_rate": 0.0002, "epoch": 4.331065759637188, "step": 1910}, {"loss": 1.8431, "grad_norm": 0.7173832654953003, "learning_rate": 0.0002, "epoch": 4.35374149659864, "step": 1920}, {"loss": 1.83, "grad_norm": 0.8341949582099915, "learning_rate": 0.0002, "epoch": 4.37641723356009, "step": 1930}, {"loss": 1.7791, "grad_norm": 0.7944530844688416, "learning_rate": 0.0002, "epoch": 4.399092970521542, "step": 1940}, {"loss": 1.8161, "grad_norm": 0.7409125566482544, "learning_rate": 0.0002, "epoch": 4.421768707482993, "step": 1950}, {"loss": 1.7994, "grad_norm": 0.9393547773361206, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 1960}, {"loss": 1.8132, "grad_norm": 0.8280852437019348, "learning_rate": 0.0002, "epoch": 4.467120181405896, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.7249669432640076, "learning_rate": 0.0002, "epoch": 4.489795918367347, "step": 1980}, {"loss": 1.7846, "grad_norm": 0.7199458479881287, "learning_rate": 0.0002, "epoch": 4.512471655328798, "step": 1990}, {"loss": 1.7851, "grad_norm": 0.8454286456108093, "learning_rate": 0.0002, "epoch": 4.535147392290249, "step": 2000}, {"loss": 1.8673, "grad_norm": 0.7305117249488831, "learning_rate": 0.0002, "epoch": 4.557823129251701, "step": 2010}, {"loss": 1.8234, "grad_norm": 0.7819753885269165, "learning_rate": 0.0002, "epoch": 4.580498866213152, "step": 2020}, {"loss": 1.7972, "grad_norm": 0.8010004162788391, "learning_rate": 0.0002, "epoch": 4.603174603174603, "step": 2030}, {"loss": 1.7921, "grad_norm": 0.7850886583328247, "learning_rate": 0.0002, "epoch": 4.625850340136054, "step": 2040}, {"loss": 1.9024, "grad_norm": 0.7433428168296814, "learning_rate": 0.0002, "epoch": 4.648526077097506, "step": 2050}, {"loss": 1.8155, "grad_norm": 0.7518399357795715, "learning_rate": 0.0002, "epoch": 4.671201814058957, "step": 2060}, {"loss": 1.7924, "grad_norm": 0.7282178997993469, "learning_rate": 0.0002, "epoch": 4.6938775510204085, "step": 2070}, {"loss": 1.8144, "grad_norm": 0.6950743794441223, "learning_rate": 0.0002, "epoch": 4.71655328798186, "step": 2080}, {"loss": 1.7757, "grad_norm": 0.761830747127533, "learning_rate": 0.0002, "epoch": 4.73922902494331, "step": 2090}, {"loss": 1.8254, "grad_norm": 0.8068088293075562, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 2100}, {"loss": 1.8092, "grad_norm": 0.7580569386482239, "learning_rate": 0.0002, "epoch": 4.784580498866213, "step": 2110}, {"loss": 1.8144, "grad_norm": 0.9428760409355164, "learning_rate": 0.0002, "epoch": 4.807256235827665, "step": 2120}, {"loss": 1.829, "grad_norm": 0.7057452201843262, "learning_rate": 0.0002, "epoch": 4.829931972789115, "step": 2130}, {"loss": 1.7603, "grad_norm": 0.763630211353302, "learning_rate": 0.0002, "epoch": 4.852607709750567, "step": 2140}, {"loss": 1.8348, "grad_norm": 0.7298727035522461, "learning_rate": 0.0002, "epoch": 4.875283446712018, "step": 2150}, {"loss": 1.8652, "grad_norm": 0.7376292943954468, "learning_rate": 0.0002, "epoch": 4.8979591836734695, "step": 2160}, {"loss": 1.742, "grad_norm": 0.8528493642807007, "learning_rate": 0.0002, "epoch": 4.920634920634921, "step": 2170}, {"loss": 1.7706, "grad_norm": 0.7652511596679688, "learning_rate": 0.0002, "epoch": 4.943310657596372, "step": 2180}, {"loss": 1.7994, "grad_norm": 0.7602494955062866, "learning_rate": 0.0002, "epoch": 4.965986394557823, "step": 2190}, {"loss": 1.8278, "grad_norm": 0.758914053440094, "learning_rate": 0.0002, "epoch": 4.988662131519274, "step": 2200}, {"eval_loss": 2.083296298980713, "eval_runtime": 20.2201, "eval_samples_per_second": 25.074, "eval_steps_per_second": 3.165, "epoch": 5.0, "step": 2205}, {"loss": 1.7003, "grad_norm": 0.6990088224411011, "learning_rate": 0.0002, "epoch": 5.011337868480726, "step": 2210}, {"loss": 1.6451, "grad_norm": 0.7627242803573608, "learning_rate": 0.0002, "epoch": 5.034013605442177, "step": 2220}, {"loss": 1.7713, "grad_norm": 0.8485700488090515, "learning_rate": 0.0002, "epoch": 5.0566893424036286, "step": 2230}, {"loss": 1.7143, "grad_norm": 0.8308766484260559, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 2240}, {"loss": 1.7318, "grad_norm": 0.825551450252533, "learning_rate": 0.0002, "epoch": 5.1020408163265305, "step": 2250}, {"loss": 1.7005, "grad_norm": 0.8571329116821289, "learning_rate": 0.0002, "epoch": 5.124716553287982, "step": 2260}, {"loss": 1.7304, "grad_norm": 0.7965090274810791, "learning_rate": 0.0002, "epoch": 5.147392290249433, "step": 2270}, {"loss": 1.713, "grad_norm": 0.7926788330078125, "learning_rate": 0.0002, "epoch": 5.170068027210885, "step": 2280}, {"loss": 1.7778, "grad_norm": 0.8400393128395081, "learning_rate": 0.0002, "epoch": 5.192743764172335, "step": 2290}, {"loss": 1.7163, "grad_norm": 0.9096018075942993, "learning_rate": 0.0002, "epoch": 5.215419501133787, "step": 2300}, {"loss": 1.7158, "grad_norm": 0.8156160712242126, "learning_rate": 0.0002, "epoch": 5.238095238095238, "step": 2310}, {"loss": 1.6941, "grad_norm": 0.8418492674827576, "learning_rate": 0.0002, "epoch": 5.26077097505669, "step": 2320}, {"loss": 1.7604, "grad_norm": 0.7994900345802307, "learning_rate": 0.0002, "epoch": 5.283446712018141, "step": 2330}, {"loss": 1.6869, "grad_norm": 0.9292947053909302, "learning_rate": 0.0002, "epoch": 5.3061224489795915, "step": 2340}, {"loss": 1.8213, "grad_norm": 0.8896278142929077, "learning_rate": 0.0002, "epoch": 5.328798185941043, "step": 2350}, {"loss": 1.6681, "grad_norm": 0.9424751400947571, "learning_rate": 0.0002, "epoch": 5.351473922902494, "step": 2360}, {"loss": 1.7032, "grad_norm": 0.8420370817184448, "learning_rate": 0.0002, "epoch": 5.374149659863946, "step": 2370}, {"loss": 1.665, "grad_norm": 0.8460527658462524, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 2380}, {"loss": 1.6443, "grad_norm": 1.0171664953231812, "learning_rate": 0.0002, "epoch": 5.419501133786848, "step": 2390}, {"loss": 1.744, "grad_norm": 0.9017615914344788, "learning_rate": 0.0002, "epoch": 5.442176870748299, "step": 2400}, {"loss": 1.7046, "grad_norm": 0.8650022745132446, "learning_rate": 0.0002, "epoch": 5.464852607709751, "step": 2410}, {"loss": 1.722, "grad_norm": 0.8252440094947815, "learning_rate": 0.0002, "epoch": 5.487528344671202, "step": 2420}, {"loss": 1.7473, "grad_norm": 0.9326912760734558, "learning_rate": 0.0002, "epoch": 5.510204081632653, "step": 2430}, {"loss": 1.7738, "grad_norm": 0.8913490772247314, "learning_rate": 0.0002, "epoch": 5.532879818594104, "step": 2440}, {"loss": 1.7197, "grad_norm": 0.8588173985481262, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 2450}, {"loss": 1.679, "grad_norm": 0.8768706321716309, "learning_rate": 0.0002, "epoch": 5.578231292517007, "step": 2460}, {"loss": 1.7005, "grad_norm": 0.9485137462615967, "learning_rate": 0.0002, "epoch": 5.600907029478458, "step": 2470}, {"loss": 1.6494, "grad_norm": 0.8462285399436951, "learning_rate": 0.0002, "epoch": 5.62358276643991, "step": 2480}, {"loss": 1.7199, "grad_norm": 0.8390767574310303, "learning_rate": 0.0002, "epoch": 5.646258503401361, "step": 2490}, {"loss": 1.7819, "grad_norm": 0.9576658010482788, "learning_rate": 0.0002, "epoch": 5.668934240362812, "step": 2500}, {"loss": 1.6684, "grad_norm": 0.9528797268867493, "learning_rate": 0.0002, "epoch": 5.691609977324263, "step": 2510}, {"loss": 1.6769, "grad_norm": 1.19835364818573, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2520}, {"loss": 1.6454, "grad_norm": 0.8822922110557556, "learning_rate": 0.0002, "epoch": 5.736961451247166, "step": 2530}, {"loss": 1.7769, "grad_norm": 0.9812421202659607, "learning_rate": 0.0002, "epoch": 5.759637188208616, "step": 2540}, {"loss": 1.6781, "grad_norm": 0.9178062677383423, "learning_rate": 0.0002, "epoch": 5.782312925170068, "step": 2550}, {"loss": 1.685, "grad_norm": 1.0217735767364502, "learning_rate": 0.0002, "epoch": 5.804988662131519, "step": 2560}, {"loss": 1.722, "grad_norm": 0.8624402284622192, "learning_rate": 0.0002, "epoch": 5.827664399092971, "step": 2570}, {"loss": 1.6976, "grad_norm": 0.8644461035728455, "learning_rate": 0.0002, "epoch": 5.850340136054422, "step": 2580}, {"loss": 1.7498, "grad_norm": 0.8437298536300659, "learning_rate": 0.0002, "epoch": 5.8730158730158735, "step": 2590}, {"loss": 1.7577, "grad_norm": 0.7952260971069336, "learning_rate": 0.0002, "epoch": 5.895691609977324, "step": 2600}, {"loss": 1.743, "grad_norm": 0.9374001622200012, "learning_rate": 0.0002, "epoch": 5.918367346938775, "step": 2610}, {"loss": 1.7689, "grad_norm": 0.8157470226287842, "learning_rate": 0.0002, "epoch": 5.941043083900227, "step": 2620}, {"loss": 1.6757, "grad_norm": 0.8078457713127136, "learning_rate": 0.0002, "epoch": 5.963718820861678, "step": 2630}, {"loss": 1.7059, "grad_norm": 0.8926942348480225, "learning_rate": 0.0002, "epoch": 5.986394557823129, "step": 2640}, {"eval_loss": 2.117249011993408, "eval_runtime": 14.7114, "eval_samples_per_second": 34.463, "eval_steps_per_second": 4.35, "epoch": 6.0, "step": 2646}, {"loss": 1.6913, "grad_norm": 0.866627037525177, "learning_rate": 0.0002, "epoch": 6.00907029478458, "step": 2650}, {"loss": 1.5567, "grad_norm": 0.9636926054954529, "learning_rate": 0.0002, "epoch": 6.031746031746032, "step": 2660}, {"loss": 1.5617, "grad_norm": 0.9994971752166748, "learning_rate": 0.0002, "epoch": 6.054421768707483, "step": 2670}, {"loss": 1.5724, "grad_norm": 0.886381983757019, "learning_rate": 0.0002, "epoch": 6.0770975056689345, "step": 2680}, {"loss": 1.7039, "grad_norm": 0.8596521615982056, "learning_rate": 0.0002, "epoch": 6.099773242630386, "step": 2690}, {"loss": 1.6268, "grad_norm": 1.0034657716751099, "learning_rate": 0.0002, "epoch": 6.122448979591836, "step": 2700}, {"loss": 1.5562, "grad_norm": 0.8961828947067261, "learning_rate": 0.0002, "epoch": 6.145124716553288, "step": 2710}, {"loss": 1.5924, "grad_norm": 0.9269746541976929, "learning_rate": 0.0002, "epoch": 6.167800453514739, "step": 2720}, {"loss": 1.6335, "grad_norm": 0.9434096217155457, "learning_rate": 0.0002, "epoch": 6.190476190476191, "step": 2730}, {"loss": 1.6189, "grad_norm": 0.9009348750114441, "learning_rate": 0.0002, "epoch": 6.213151927437642, "step": 2740}, {"loss": 1.6322, "grad_norm": 1.0572991371154785, "learning_rate": 0.0002, "epoch": 6.235827664399093, "step": 2750}, {"loss": 1.6675, "grad_norm": 0.996005654335022, "learning_rate": 0.0002, "epoch": 6.258503401360544, "step": 2760}, {"loss": 1.5836, "grad_norm": 0.8857560753822327, "learning_rate": 0.0002, "epoch": 6.2811791383219955, "step": 2770}, {"loss": 1.6365, "grad_norm": 0.9528077840805054, "learning_rate": 0.0002, "epoch": 6.303854875283447, "step": 2780}, {"loss": 1.636, "grad_norm": 0.8799851536750793, "learning_rate": 0.0002, "epoch": 6.326530612244898, "step": 2790}, {"loss": 1.6227, "grad_norm": 0.9826605916023254, "learning_rate": 0.0002, "epoch": 6.349206349206349, "step": 2800}, {"loss": 1.5672, "grad_norm": 1.0445730686187744, "learning_rate": 0.0002, "epoch": 6.3718820861678, "step": 2810}, {"loss": 1.6476, "grad_norm": 1.0388351678848267, "learning_rate": 0.0002, "epoch": 6.394557823129252, "step": 2820}, {"loss": 1.5719, "grad_norm": 0.9709566831588745, "learning_rate": 0.0002, "epoch": 6.417233560090703, "step": 2830}, {"loss": 1.6654, "grad_norm": 0.9056107401847839, "learning_rate": 0.0002, "epoch": 6.4399092970521545, "step": 2840}, {"loss": 1.6482, "grad_norm": 1.0178978443145752, "learning_rate": 0.0002, "epoch": 6.462585034013605, "step": 2850}, {"loss": 1.5917, "grad_norm": 0.9525032639503479, "learning_rate": 0.0002, "epoch": 6.4852607709750565, "step": 2860}, {"loss": 1.6261, "grad_norm": 0.9033688306808472, "learning_rate": 0.0002, "epoch": 6.507936507936508, "step": 2870}, {"loss": 1.5718, "grad_norm": 0.9108299016952515, "learning_rate": 0.0002, "epoch": 6.530612244897959, "step": 2880}, {"loss": 1.6426, "grad_norm": 0.961175262928009, "learning_rate": 0.0002, "epoch": 6.553287981859411, "step": 2890}, {"loss": 1.5573, "grad_norm": 0.8874754309654236, "learning_rate": 0.0002, "epoch": 6.575963718820862, "step": 2900}, {"loss": 1.6003, "grad_norm": 0.9972973465919495, "learning_rate": 0.0002, "epoch": 6.598639455782313, "step": 2910}, {"loss": 1.5926, "grad_norm": 0.9361436367034912, "learning_rate": 0.0002, "epoch": 6.621315192743764, "step": 2920}, {"loss": 1.6858, "grad_norm": 0.9379439353942871, "learning_rate": 0.0002, "epoch": 6.6439909297052155, "step": 2930}, {"loss": 1.6558, "grad_norm": 0.9896080493927002, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 2940}, {"loss": 1.6045, "grad_norm": 1.0048781633377075, "learning_rate": 0.0002, "epoch": 6.6893424036281175, "step": 2950}, {"loss": 1.6269, "grad_norm": 0.9979572892189026, "learning_rate": 0.0002, "epoch": 6.712018140589569, "step": 2960}, {"loss": 1.6663, "grad_norm": 0.9461016654968262, "learning_rate": 0.0002, "epoch": 6.73469387755102, "step": 2970}, {"loss": 1.6948, "grad_norm": 0.9460693597793579, "learning_rate": 0.0002, "epoch": 6.757369614512472, "step": 2980}, {"loss": 1.5654, "grad_norm": 0.9581215977668762, "learning_rate": 0.0002, "epoch": 6.780045351473923, "step": 2990}, {"loss": 1.6814, "grad_norm": 0.8998590111732483, "learning_rate": 0.0002, "epoch": 6.802721088435375, "step": 3000}, {"loss": 1.6812, "grad_norm": 0.9226139187812805, "learning_rate": 0.0002, "epoch": 6.825396825396825, "step": 3010}, {"loss": 1.6407, "grad_norm": 0.9358452558517456, "learning_rate": 0.0002, "epoch": 6.8480725623582765, "step": 3020}, {"loss": 1.6654, "grad_norm": 1.1139743328094482, "learning_rate": 0.0002, "epoch": 6.870748299319728, "step": 3030}, {"loss": 1.6326, "grad_norm": 0.9886892437934875, "learning_rate": 0.0002, "epoch": 6.893424036281179, "step": 3040}, {"loss": 1.6329, "grad_norm": 0.885146975517273, "learning_rate": 0.0002, "epoch": 6.91609977324263, "step": 3050}, {"loss": 1.6473, "grad_norm": 0.9668041467666626, "learning_rate": 0.0002, "epoch": 6.938775510204081, "step": 3060}, {"loss": 1.5385, "grad_norm": 0.9532362222671509, "learning_rate": 0.0002, "epoch": 6.961451247165533, "step": 3070}, {"loss": 1.5744, "grad_norm": 1.0338832139968872, "learning_rate": 0.0002, "epoch": 6.984126984126984, "step": 3080}, {"eval_loss": 2.155296564102173, "eval_runtime": 14.7454, "eval_samples_per_second": 34.384, "eval_steps_per_second": 4.34, "epoch": 7.0, "step": 3087}, {"loss": 1.5442, "grad_norm": 0.979254424571991, "learning_rate": 0.0002, "epoch": 7.006802721088436, "step": 3090}, {"loss": 1.5831, "grad_norm": 1.1125438213348389, "learning_rate": 0.0002, "epoch": 7.029478458049887, "step": 3100}, {"loss": 1.5674, "grad_norm": 1.080736517906189, "learning_rate": 0.0002, "epoch": 7.052154195011338, "step": 3110}, {"loss": 1.4347, "grad_norm": 1.0562000274658203, "learning_rate": 0.0002, "epoch": 7.074829931972789, "step": 3120}, {"loss": 1.4841, "grad_norm": 1.022085428237915, "learning_rate": 0.0002, "epoch": 7.09750566893424, "step": 3130}, {"loss": 1.5087, "grad_norm": 1.0203129053115845, "learning_rate": 0.0002, "epoch": 7.120181405895692, "step": 3140}, {"loss": 1.546, "grad_norm": 1.0613057613372803, "learning_rate": 0.0002, "epoch": 7.142857142857143, "step": 3150}, {"loss": 1.5131, "grad_norm": 1.2065532207489014, "learning_rate": 0.0002, "epoch": 7.165532879818594, "step": 3160}, {"loss": 1.5023, "grad_norm": 1.0489833354949951, "learning_rate": 0.0002, "epoch": 7.188208616780045, "step": 3170}, {"loss": 1.5058, "grad_norm": 1.1445655822753906, "learning_rate": 0.0002, "epoch": 7.210884353741497, "step": 3180}, {"loss": 1.5307, "grad_norm": 1.0830376148223877, "learning_rate": 0.0002, "epoch": 7.233560090702948, "step": 3190}, {"loss": 1.5298, "grad_norm": 1.152576208114624, "learning_rate": 0.0002, "epoch": 7.2562358276643995, "step": 3200}, {"loss": 1.4823, "grad_norm": 1.2006800174713135, "learning_rate": 0.0002, "epoch": 7.27891156462585, "step": 3210}, {"loss": 1.4567, "grad_norm": 1.0562821626663208, "learning_rate": 0.0002, "epoch": 7.301587301587301, "step": 3220}, {"loss": 1.5729, "grad_norm": 1.0565770864486694, "learning_rate": 0.0002, "epoch": 7.324263038548753, "step": 3230}, {"loss": 1.5353, "grad_norm": 1.0950746536254883, "learning_rate": 0.0002, "epoch": 7.346938775510204, "step": 3240}, {"loss": 1.4982, "grad_norm": 1.0843168497085571, "learning_rate": 0.0002, "epoch": 7.369614512471656, "step": 3250}, {"loss": 1.5202, "grad_norm": 1.1089458465576172, "learning_rate": 0.0002, "epoch": 7.392290249433106, "step": 3260}, {"loss": 1.5477, "grad_norm": 1.1431858539581299, "learning_rate": 0.0002, "epoch": 7.414965986394558, "step": 3270}, {"loss": 1.477, "grad_norm": 1.1291835308074951, "learning_rate": 0.0002, "epoch": 7.437641723356009, "step": 3280}, {"loss": 1.4595, "grad_norm": 1.1871713399887085, "learning_rate": 0.0002, "epoch": 7.4603174603174605, "step": 3290}, {"loss": 1.5366, "grad_norm": 1.0716718435287476, "learning_rate": 0.0002, "epoch": 7.482993197278912, "step": 3300}, {"loss": 1.5334, "grad_norm": 1.0762914419174194, "learning_rate": 0.0002, "epoch": 7.505668934240363, "step": 3310}, {"loss": 1.5113, "grad_norm": 1.0819021463394165, "learning_rate": 0.0002, "epoch": 7.528344671201814, "step": 3320}, {"loss": 1.4996, "grad_norm": 1.1713459491729736, "learning_rate": 0.0002, "epoch": 7.551020408163265, "step": 3330}, {"loss": 1.5054, "grad_norm": 1.0692853927612305, "learning_rate": 0.0002, "epoch": 7.573696145124717, "step": 3340}, {"loss": 1.5046, "grad_norm": 1.2106385231018066, "learning_rate": 0.0002, "epoch": 7.596371882086168, "step": 3350}, {"loss": 1.5189, "grad_norm": 1.116910457611084, "learning_rate": 0.0002, "epoch": 7.619047619047619, "step": 3360}, {"loss": 1.505, "grad_norm": 1.105789303779602, "learning_rate": 0.0002, "epoch": 7.64172335600907, "step": 3370}, {"loss": 1.5334, "grad_norm": 1.1199218034744263, "learning_rate": 0.0002, "epoch": 7.6643990929705215, "step": 3380}, {"loss": 1.4746, "grad_norm": 1.056763768196106, "learning_rate": 0.0002, "epoch": 7.687074829931973, "step": 3390}, {"loss": 1.5497, "grad_norm": 1.0940561294555664, "learning_rate": 0.0002, "epoch": 7.709750566893424, "step": 3400}, {"loss": 1.5386, "grad_norm": 1.100599765777588, "learning_rate": 0.0002, "epoch": 7.732426303854876, "step": 3410}, {"loss": 1.5392, "grad_norm": 1.2683327198028564, "learning_rate": 0.0002, "epoch": 7.755102040816326, "step": 3420}, {"loss": 1.5954, "grad_norm": 1.140558123588562, "learning_rate": 0.0002, "epoch": 7.777777777777778, "step": 3430}, {"loss": 1.4685, "grad_norm": 1.096411943435669, "learning_rate": 0.0002, "epoch": 7.800453514739229, "step": 3440}, {"loss": 1.4702, "grad_norm": 1.1843889951705933, "learning_rate": 0.0002, "epoch": 7.8231292517006805, "step": 3450}, {"loss": 1.5122, "grad_norm": 1.1010003089904785, "learning_rate": 0.0002, "epoch": 7.845804988662131, "step": 3460}, {"loss": 1.5447, "grad_norm": 1.0672950744628906, "learning_rate": 0.0002, "epoch": 7.8684807256235825, "step": 3470}, {"loss": 1.5641, "grad_norm": 1.0926506519317627, "learning_rate": 0.0002, "epoch": 7.891156462585034, "step": 3480}, {"loss": 1.5274, "grad_norm": 1.0815032720565796, "learning_rate": 0.0002, "epoch": 7.913832199546485, "step": 3490}, {"loss": 1.5018, "grad_norm": 1.2033658027648926, "learning_rate": 0.0002, "epoch": 7.936507936507937, "step": 3500}, {"loss": 1.5767, "grad_norm": 1.131916880607605, "learning_rate": 0.0002, "epoch": 7.959183673469388, "step": 3510}, {"loss": 1.5814, "grad_norm": 1.0860273838043213, "learning_rate": 0.0002, "epoch": 7.981859410430839, "step": 3520}]}