diff --git a/.gitattributes b/.gitattributes
index d13c36e40ae6cbc37de042069d05a0241a63d314..82ed3f8dbfc0ed9a81e12460b22dfd5463cb0b28 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -789,3 +789,12 @@ gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d
 gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-fixed_num-p-0.03-num-200-sd-42/checkpoint-40/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-fixed_num-p-0.03-num-200-sd-42/checkpoint-81/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-fixed_num-p-0.03-num-200-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8b971d2d3728c073474494167a8733e971e412cf
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a973c78893d5c5f843df8157e4f5f5d58bd4afd185281397acca339964604cf
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..aca91d18460f5dd0046c61b1bcc21240a46ec829
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49c6f69d70eebb972d5f2aea734807cdc237ada8f0fd2bcc1d42cf91146986ff
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c1e95bd59d08cef07e5077d41462b847bfcb154
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27572929ff2cbf856253f58f8e5059eee8bb373275a4d66ee2157c75df2e9641
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e43d3a86d9e002f165c21f20848463d0d9b4a483
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90c62348675de8dfb807cfdc1360f392af5cab7689636d8eb0d4df84ca06a498
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2d4595f2f1663ea0bd12a695c82cde8c2bf8a19e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cc0c19e0ddc643f135d0fdd597ab58ee08d8a244658aab7ba0b5d38110da455
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..81f785078fede5007798b307e9120007bd9cd895
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/trainer_state.json
@@ -0,0 +1,974 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 2.998857142857143,
+  "eval_steps": 10,
+  "global_step": 1312,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.741491613696e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1312/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3f5e59110eabce72d08ce0237a3c9fc0572c7edb
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1750e91eb527bd511e256a80ba1d623c41bee957c5f1b88c3a3feede9651275f
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..47aeeaf29ac5ae512dc6432b8bc3084c63fb8815
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78544d3cfcb349ba50a10fed20793113a10afd6e720e8d0e5e84cedf15ed694d
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..99ce54da6e759399277a9ae48dc5955dce873c26
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c4cd5653f9fb65ce60bb97044c048a65fcc5c435bf82c98d80a311e76d97d93
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..178b869d2471e46c12c962fa4d3f49da80ade0ba
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c637c522eb7be4f20db9bf30934ffbcff195e29b30bd9e03e47a7a09113597c
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0d1dff582f37765ea5b94f069e83da291fb056ed
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/trainer_state.json
@@ -0,0 +1,1290 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 1750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    },
+    {
+      "epoch": 3.0171428571428573,
+      "grad_norm": 0.5184893012046814,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 1320
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.5665355920791626,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0628571428571427,
+      "grad_norm": 0.6601403951644897,
+      "learning_rate": 0.0002,
+      "loss": 1.3741,
+      "step": 1340
+    },
+    {
+      "epoch": 3.085714285714286,
+      "grad_norm": 0.6921621561050415,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 1350
+    },
+    {
+      "epoch": 3.1085714285714285,
+      "grad_norm": 0.6406348943710327,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 1360
+    },
+    {
+      "epoch": 3.1314285714285712,
+      "grad_norm": 0.5814554691314697,
+      "learning_rate": 0.0002,
+      "loss": 1.3563,
+      "step": 1370
+    },
+    {
+      "epoch": 3.1542857142857144,
+      "grad_norm": 0.683325469493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 1380
+    },
+    {
+      "epoch": 3.177142857142857,
+      "grad_norm": 0.6686155200004578,
+      "learning_rate": 0.0002,
+      "loss": 1.4106,
+      "step": 1390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8159713745117188,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 1400
+    },
+    {
+      "epoch": 3.222857142857143,
+      "grad_norm": 0.646216094493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4279,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2457142857142856,
+      "grad_norm": 0.7323529720306396,
+      "learning_rate": 0.0002,
+      "loss": 1.4232,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2685714285714287,
+      "grad_norm": 0.689349353313446,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2914285714285714,
+      "grad_norm": 0.727894127368927,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 1440
+    },
+    {
+      "epoch": 3.314285714285714,
+      "grad_norm": 0.6921590566635132,
+      "learning_rate": 0.0002,
+      "loss": 1.4,
+      "step": 1450
+    },
+    {
+      "epoch": 3.337142857142857,
+      "grad_norm": 0.6176243424415588,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 1460
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.9006354212760925,
+      "learning_rate": 0.0002,
+      "loss": 1.4323,
+      "step": 1470
+    },
+    {
+      "epoch": 3.382857142857143,
+      "grad_norm": 0.8145929574966431,
+      "learning_rate": 0.0002,
+      "loss": 1.4353,
+      "step": 1480
+    },
+    {
+      "epoch": 3.4057142857142857,
+      "grad_norm": 0.6640016436576843,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.7266780138015747,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4514285714285715,
+      "grad_norm": 0.9351356029510498,
+      "learning_rate": 0.0002,
+      "loss": 1.4108,
+      "step": 1510
+    },
+    {
+      "epoch": 3.474285714285714,
+      "grad_norm": 0.675645649433136,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4971428571428573,
+      "grad_norm": 0.761472225189209,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 1530
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.6653069257736206,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5428571428571427,
+      "grad_norm": 0.667412519454956,
+      "learning_rate": 0.0002,
+      "loss": 1.4686,
+      "step": 1550
+    },
+    {
+      "epoch": 3.565714285714286,
+      "grad_norm": 0.6395593881607056,
+      "learning_rate": 0.0002,
+      "loss": 1.4241,
+      "step": 1560
+    },
+    {
+      "epoch": 3.5885714285714285,
+      "grad_norm": 0.7588621377944946,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1570
+    },
+    {
+      "epoch": 3.611428571428571,
+      "grad_norm": 0.6206456422805786,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6342857142857143,
+      "grad_norm": 0.7591291666030884,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 1590
+    },
+    {
+      "epoch": 3.657142857142857,
+      "grad_norm": 0.6476313471794128,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 1600
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.6731392741203308,
+      "learning_rate": 0.0002,
+      "loss": 1.4598,
+      "step": 1610
+    },
+    {
+      "epoch": 3.702857142857143,
+      "grad_norm": 0.725190281867981,
+      "learning_rate": 0.0002,
+      "loss": 1.4225,
+      "step": 1620
+    },
+    {
+      "epoch": 3.725714285714286,
+      "grad_norm": 0.6720049977302551,
+      "learning_rate": 0.0002,
+      "loss": 1.4525,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7485714285714287,
+      "grad_norm": 0.6301007270812988,
+      "learning_rate": 0.0002,
+      "loss": 1.429,
+      "step": 1640
+    },
+    {
+      "epoch": 3.7714285714285714,
+      "grad_norm": 0.715893566608429,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 1650
+    },
+    {
+      "epoch": 3.7942857142857145,
+      "grad_norm": 0.7539359927177429,
+      "learning_rate": 0.0002,
+      "loss": 1.3624,
+      "step": 1660
+    },
+    {
+      "epoch": 3.817142857142857,
+      "grad_norm": 0.6658543348312378,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 1670
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.7019526958465576,
+      "learning_rate": 0.0002,
+      "loss": 1.3934,
+      "step": 1680
+    },
+    {
+      "epoch": 3.862857142857143,
+      "grad_norm": 0.6517802476882935,
+      "learning_rate": 0.0002,
+      "loss": 1.4436,
+      "step": 1690
+    },
+    {
+      "epoch": 3.8857142857142857,
+      "grad_norm": 0.7617332935333252,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 3.9085714285714284,
+      "grad_norm": 0.6919480562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9314285714285715,
+      "grad_norm": 0.6987943053245544,
+      "learning_rate": 0.0002,
+      "loss": 1.4317,
+      "step": 1720
+    },
+    {
+      "epoch": 3.954285714285714,
+      "grad_norm": 0.7062228918075562,
+      "learning_rate": 0.0002,
+      "loss": 1.4704,
+      "step": 1730
+    },
+    {
+      "epoch": 3.977142857142857,
+      "grad_norm": 0.6769542098045349,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 1740
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6832144260406494,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 1750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9474865198135376,
+      "eval_runtime": 111.288,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1750
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 8.988655484928e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-1750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..824b95c2b7e49523c0756d62bc3cb36d3c13af69
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1b156501fe3efd5331277d2b001a479c2e4b7a28af955567f4f6646a0ddb0c7
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..21cf4c8eacdd808613116807ba97a81e05d58a0d
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f02e78e00308b334c9278f652b9d7c9629f126f03ff89e55da48446b1b81ac28
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e2cfeb28991d621d38b81314564cb90650097838
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b926dd98f2307096cd91a16232f4c376f4061f738c6914f96b03494e6fdc2070
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1f783b51ae79aa4985c73df8b5ca9af4e84433e9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d8e3dec179fb1345860d00fab246de4b51f381c286b4405ca26f027be2ba498
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b76bd5e6fedf2a32b1e8f50095c67b0f6053da50
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/trainer_state.json
@@ -0,0 +1,1599 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 4.998857142857143,
+  "eval_steps": 10,
+  "global_step": 2187,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    },
+    {
+      "epoch": 3.0171428571428573,
+      "grad_norm": 0.5184893012046814,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 1320
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.5665355920791626,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0628571428571427,
+      "grad_norm": 0.6601403951644897,
+      "learning_rate": 0.0002,
+      "loss": 1.3741,
+      "step": 1340
+    },
+    {
+      "epoch": 3.085714285714286,
+      "grad_norm": 0.6921621561050415,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 1350
+    },
+    {
+      "epoch": 3.1085714285714285,
+      "grad_norm": 0.6406348943710327,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 1360
+    },
+    {
+      "epoch": 3.1314285714285712,
+      "grad_norm": 0.5814554691314697,
+      "learning_rate": 0.0002,
+      "loss": 1.3563,
+      "step": 1370
+    },
+    {
+      "epoch": 3.1542857142857144,
+      "grad_norm": 0.683325469493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 1380
+    },
+    {
+      "epoch": 3.177142857142857,
+      "grad_norm": 0.6686155200004578,
+      "learning_rate": 0.0002,
+      "loss": 1.4106,
+      "step": 1390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8159713745117188,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 1400
+    },
+    {
+      "epoch": 3.222857142857143,
+      "grad_norm": 0.646216094493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4279,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2457142857142856,
+      "grad_norm": 0.7323529720306396,
+      "learning_rate": 0.0002,
+      "loss": 1.4232,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2685714285714287,
+      "grad_norm": 0.689349353313446,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2914285714285714,
+      "grad_norm": 0.727894127368927,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 1440
+    },
+    {
+      "epoch": 3.314285714285714,
+      "grad_norm": 0.6921590566635132,
+      "learning_rate": 0.0002,
+      "loss": 1.4,
+      "step": 1450
+    },
+    {
+      "epoch": 3.337142857142857,
+      "grad_norm": 0.6176243424415588,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 1460
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.9006354212760925,
+      "learning_rate": 0.0002,
+      "loss": 1.4323,
+      "step": 1470
+    },
+    {
+      "epoch": 3.382857142857143,
+      "grad_norm": 0.8145929574966431,
+      "learning_rate": 0.0002,
+      "loss": 1.4353,
+      "step": 1480
+    },
+    {
+      "epoch": 3.4057142857142857,
+      "grad_norm": 0.6640016436576843,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.7266780138015747,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4514285714285715,
+      "grad_norm": 0.9351356029510498,
+      "learning_rate": 0.0002,
+      "loss": 1.4108,
+      "step": 1510
+    },
+    {
+      "epoch": 3.474285714285714,
+      "grad_norm": 0.675645649433136,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4971428571428573,
+      "grad_norm": 0.761472225189209,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 1530
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.6653069257736206,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5428571428571427,
+      "grad_norm": 0.667412519454956,
+      "learning_rate": 0.0002,
+      "loss": 1.4686,
+      "step": 1550
+    },
+    {
+      "epoch": 3.565714285714286,
+      "grad_norm": 0.6395593881607056,
+      "learning_rate": 0.0002,
+      "loss": 1.4241,
+      "step": 1560
+    },
+    {
+      "epoch": 3.5885714285714285,
+      "grad_norm": 0.7588621377944946,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1570
+    },
+    {
+      "epoch": 3.611428571428571,
+      "grad_norm": 0.6206456422805786,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6342857142857143,
+      "grad_norm": 0.7591291666030884,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 1590
+    },
+    {
+      "epoch": 3.657142857142857,
+      "grad_norm": 0.6476313471794128,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 1600
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.6731392741203308,
+      "learning_rate": 0.0002,
+      "loss": 1.4598,
+      "step": 1610
+    },
+    {
+      "epoch": 3.702857142857143,
+      "grad_norm": 0.725190281867981,
+      "learning_rate": 0.0002,
+      "loss": 1.4225,
+      "step": 1620
+    },
+    {
+      "epoch": 3.725714285714286,
+      "grad_norm": 0.6720049977302551,
+      "learning_rate": 0.0002,
+      "loss": 1.4525,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7485714285714287,
+      "grad_norm": 0.6301007270812988,
+      "learning_rate": 0.0002,
+      "loss": 1.429,
+      "step": 1640
+    },
+    {
+      "epoch": 3.7714285714285714,
+      "grad_norm": 0.715893566608429,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 1650
+    },
+    {
+      "epoch": 3.7942857142857145,
+      "grad_norm": 0.7539359927177429,
+      "learning_rate": 0.0002,
+      "loss": 1.3624,
+      "step": 1660
+    },
+    {
+      "epoch": 3.817142857142857,
+      "grad_norm": 0.6658543348312378,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 1670
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.7019526958465576,
+      "learning_rate": 0.0002,
+      "loss": 1.3934,
+      "step": 1680
+    },
+    {
+      "epoch": 3.862857142857143,
+      "grad_norm": 0.6517802476882935,
+      "learning_rate": 0.0002,
+      "loss": 1.4436,
+      "step": 1690
+    },
+    {
+      "epoch": 3.8857142857142857,
+      "grad_norm": 0.7617332935333252,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 3.9085714285714284,
+      "grad_norm": 0.6919480562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9314285714285715,
+      "grad_norm": 0.6987943053245544,
+      "learning_rate": 0.0002,
+      "loss": 1.4317,
+      "step": 1720
+    },
+    {
+      "epoch": 3.954285714285714,
+      "grad_norm": 0.7062228918075562,
+      "learning_rate": 0.0002,
+      "loss": 1.4704,
+      "step": 1730
+    },
+    {
+      "epoch": 3.977142857142857,
+      "grad_norm": 0.6769542098045349,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 1740
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6832144260406494,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 1750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9474865198135376,
+      "eval_runtime": 111.288,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1750
+    },
+    {
+      "epoch": 4.022857142857143,
+      "grad_norm": 1.064110279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 1760
+    },
+    {
+      "epoch": 4.045714285714285,
+      "grad_norm": 0.8380683660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.2013,
+      "step": 1770
+    },
+    {
+      "epoch": 4.0685714285714285,
+      "grad_norm": 1.1863020658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 1780
+    },
+    {
+      "epoch": 4.091428571428572,
+      "grad_norm": 1.0128898620605469,
+      "learning_rate": 0.0002,
+      "loss": 1.2499,
+      "step": 1790
+    },
+    {
+      "epoch": 4.114285714285714,
+      "grad_norm": 0.9221312403678894,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 1800
+    },
+    {
+      "epoch": 4.137142857142857,
+      "grad_norm": 1.1298727989196777,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 1810
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 0.8854547739028931,
+      "learning_rate": 0.0002,
+      "loss": 1.1491,
+      "step": 1820
+    },
+    {
+      "epoch": 4.182857142857143,
+      "grad_norm": 0.8920808434486389,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1830
+    },
+    {
+      "epoch": 4.2057142857142855,
+      "grad_norm": 0.913244366645813,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 1840
+    },
+    {
+      "epoch": 4.228571428571429,
+      "grad_norm": 0.908831000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1850
+    },
+    {
+      "epoch": 4.251428571428572,
+      "grad_norm": 1.0223685503005981,
+      "learning_rate": 0.0002,
+      "loss": 1.1653,
+      "step": 1860
+    },
+    {
+      "epoch": 4.274285714285714,
+      "grad_norm": 0.9771921634674072,
+      "learning_rate": 0.0002,
+      "loss": 1.2497,
+      "step": 1870
+    },
+    {
+      "epoch": 4.297142857142857,
+      "grad_norm": 0.9313384890556335,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 1880
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 1.0754257440567017,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 1890
+    },
+    {
+      "epoch": 4.3428571428571425,
+      "grad_norm": 0.8904672265052795,
+      "learning_rate": 0.0002,
+      "loss": 1.2286,
+      "step": 1900
+    },
+    {
+      "epoch": 4.365714285714286,
+      "grad_norm": 1.046527624130249,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 1910
+    },
+    {
+      "epoch": 4.388571428571429,
+      "grad_norm": 0.9576982855796814,
+      "learning_rate": 0.0002,
+      "loss": 1.2368,
+      "step": 1920
+    },
+    {
+      "epoch": 4.411428571428571,
+      "grad_norm": 0.9278356432914734,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 1930
+    },
+    {
+      "epoch": 4.434285714285714,
+      "grad_norm": 1.1763030290603638,
+      "learning_rate": 0.0002,
+      "loss": 1.2005,
+      "step": 1940
+    },
+    {
+      "epoch": 4.457142857142857,
+      "grad_norm": 0.9183000326156616,
+      "learning_rate": 0.0002,
+      "loss": 1.1541,
+      "step": 1950
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 1.050980806350708,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 1960
+    },
+    {
+      "epoch": 4.502857142857143,
+      "grad_norm": 0.9975392818450928,
+      "learning_rate": 0.0002,
+      "loss": 1.2133,
+      "step": 1970
+    },
+    {
+      "epoch": 4.525714285714286,
+      "grad_norm": 0.990544319152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2312,
+      "step": 1980
+    },
+    {
+      "epoch": 4.548571428571429,
+      "grad_norm": 1.004794955253601,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 1990
+    },
+    {
+      "epoch": 4.571428571428571,
+      "grad_norm": 0.9294857978820801,
+      "learning_rate": 0.0002,
+      "loss": 1.2085,
+      "step": 2000
+    },
+    {
+      "epoch": 4.594285714285714,
+      "grad_norm": 0.93436598777771,
+      "learning_rate": 0.0002,
+      "loss": 1.2874,
+      "step": 2010
+    },
+    {
+      "epoch": 4.617142857142857,
+      "grad_norm": 0.8704655766487122,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 2020
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 0.9077927470207214,
+      "learning_rate": 0.0002,
+      "loss": 1.204,
+      "step": 2030
+    },
+    {
+      "epoch": 4.662857142857143,
+      "grad_norm": 0.912987470626831,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 2040
+    },
+    {
+      "epoch": 4.685714285714286,
+      "grad_norm": 0.9740643501281738,
+      "learning_rate": 0.0002,
+      "loss": 1.2868,
+      "step": 2050
+    },
+    {
+      "epoch": 4.708571428571428,
+      "grad_norm": 1.133357048034668,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 2060
+    },
+    {
+      "epoch": 4.731428571428571,
+      "grad_norm": 0.8844527006149292,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 2070
+    },
+    {
+      "epoch": 4.7542857142857144,
+      "grad_norm": 1.0083311796188354,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 2080
+    },
+    {
+      "epoch": 4.777142857142858,
+      "grad_norm": 1.000447154045105,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 2090
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.9620300531387329,
+      "learning_rate": 0.0002,
+      "loss": 1.2313,
+      "step": 2100
+    },
+    {
+      "epoch": 4.822857142857143,
+      "grad_norm": 0.9843335151672363,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 2110
+    },
+    {
+      "epoch": 4.845714285714286,
+      "grad_norm": 0.9906681180000305,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 2120
+    },
+    {
+      "epoch": 4.868571428571428,
+      "grad_norm": 0.9544073939323425,
+      "learning_rate": 0.0002,
+      "loss": 1.2325,
+      "step": 2130
+    },
+    {
+      "epoch": 4.8914285714285715,
+      "grad_norm": 0.9392994046211243,
+      "learning_rate": 0.0002,
+      "loss": 1.284,
+      "step": 2140
+    },
+    {
+      "epoch": 4.914285714285715,
+      "grad_norm": 1.104519248008728,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2150
+    },
+    {
+      "epoch": 4.937142857142857,
+      "grad_norm": 0.9495956897735596,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 2160
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 0.9696287512779236,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 2170
+    },
+    {
+      "epoch": 4.982857142857143,
+      "grad_norm": 0.9933681488037109,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 2180
+    },
+    {
+      "epoch": 4.998857142857143,
+      "eval_loss": 2.099808692932129,
+      "eval_runtime": 111.2808,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 2187
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 1.123581935616e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2187/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3be37d4c6b6f37771b509101bfb565c07e558f89
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b06a7478304c7210822cbebd82f6796081559cfcb7ad1b552b11e755a36f17f9
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb33931affd94b426653e6fa523865dee02cf4ab
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcd93dda94ea710dc659dc0fb3903f2c404cfb9cf8e62d7bc0408424394559fb
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..160d882adb883975ecb841d9f241c6d568194e9d
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f50840728598e0c1ba5051a633704b00606fd9495de49a971236e0ce78e9f88
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6ffd52db6f1ca6e0c11191fe35d976561cf706c4
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d26be192efafe0f0679b00ecf85ec3d157ecf6e17a0beab0bc64146c7a8f4c3
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..f6e7548b48eea941b6a227f36c36a6f00df62eac
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/trainer_state.json
@@ -0,0 +1,1915 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 2625,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    },
+    {
+      "epoch": 3.0171428571428573,
+      "grad_norm": 0.5184893012046814,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 1320
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.5665355920791626,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0628571428571427,
+      "grad_norm": 0.6601403951644897,
+      "learning_rate": 0.0002,
+      "loss": 1.3741,
+      "step": 1340
+    },
+    {
+      "epoch": 3.085714285714286,
+      "grad_norm": 0.6921621561050415,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 1350
+    },
+    {
+      "epoch": 3.1085714285714285,
+      "grad_norm": 0.6406348943710327,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 1360
+    },
+    {
+      "epoch": 3.1314285714285712,
+      "grad_norm": 0.5814554691314697,
+      "learning_rate": 0.0002,
+      "loss": 1.3563,
+      "step": 1370
+    },
+    {
+      "epoch": 3.1542857142857144,
+      "grad_norm": 0.683325469493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 1380
+    },
+    {
+      "epoch": 3.177142857142857,
+      "grad_norm": 0.6686155200004578,
+      "learning_rate": 0.0002,
+      "loss": 1.4106,
+      "step": 1390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8159713745117188,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 1400
+    },
+    {
+      "epoch": 3.222857142857143,
+      "grad_norm": 0.646216094493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4279,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2457142857142856,
+      "grad_norm": 0.7323529720306396,
+      "learning_rate": 0.0002,
+      "loss": 1.4232,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2685714285714287,
+      "grad_norm": 0.689349353313446,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2914285714285714,
+      "grad_norm": 0.727894127368927,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 1440
+    },
+    {
+      "epoch": 3.314285714285714,
+      "grad_norm": 0.6921590566635132,
+      "learning_rate": 0.0002,
+      "loss": 1.4,
+      "step": 1450
+    },
+    {
+      "epoch": 3.337142857142857,
+      "grad_norm": 0.6176243424415588,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 1460
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.9006354212760925,
+      "learning_rate": 0.0002,
+      "loss": 1.4323,
+      "step": 1470
+    },
+    {
+      "epoch": 3.382857142857143,
+      "grad_norm": 0.8145929574966431,
+      "learning_rate": 0.0002,
+      "loss": 1.4353,
+      "step": 1480
+    },
+    {
+      "epoch": 3.4057142857142857,
+      "grad_norm": 0.6640016436576843,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.7266780138015747,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4514285714285715,
+      "grad_norm": 0.9351356029510498,
+      "learning_rate": 0.0002,
+      "loss": 1.4108,
+      "step": 1510
+    },
+    {
+      "epoch": 3.474285714285714,
+      "grad_norm": 0.675645649433136,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4971428571428573,
+      "grad_norm": 0.761472225189209,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 1530
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.6653069257736206,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5428571428571427,
+      "grad_norm": 0.667412519454956,
+      "learning_rate": 0.0002,
+      "loss": 1.4686,
+      "step": 1550
+    },
+    {
+      "epoch": 3.565714285714286,
+      "grad_norm": 0.6395593881607056,
+      "learning_rate": 0.0002,
+      "loss": 1.4241,
+      "step": 1560
+    },
+    {
+      "epoch": 3.5885714285714285,
+      "grad_norm": 0.7588621377944946,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1570
+    },
+    {
+      "epoch": 3.611428571428571,
+      "grad_norm": 0.6206456422805786,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6342857142857143,
+      "grad_norm": 0.7591291666030884,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 1590
+    },
+    {
+      "epoch": 3.657142857142857,
+      "grad_norm": 0.6476313471794128,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 1600
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.6731392741203308,
+      "learning_rate": 0.0002,
+      "loss": 1.4598,
+      "step": 1610
+    },
+    {
+      "epoch": 3.702857142857143,
+      "grad_norm": 0.725190281867981,
+      "learning_rate": 0.0002,
+      "loss": 1.4225,
+      "step": 1620
+    },
+    {
+      "epoch": 3.725714285714286,
+      "grad_norm": 0.6720049977302551,
+      "learning_rate": 0.0002,
+      "loss": 1.4525,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7485714285714287,
+      "grad_norm": 0.6301007270812988,
+      "learning_rate": 0.0002,
+      "loss": 1.429,
+      "step": 1640
+    },
+    {
+      "epoch": 3.7714285714285714,
+      "grad_norm": 0.715893566608429,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 1650
+    },
+    {
+      "epoch": 3.7942857142857145,
+      "grad_norm": 0.7539359927177429,
+      "learning_rate": 0.0002,
+      "loss": 1.3624,
+      "step": 1660
+    },
+    {
+      "epoch": 3.817142857142857,
+      "grad_norm": 0.6658543348312378,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 1670
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.7019526958465576,
+      "learning_rate": 0.0002,
+      "loss": 1.3934,
+      "step": 1680
+    },
+    {
+      "epoch": 3.862857142857143,
+      "grad_norm": 0.6517802476882935,
+      "learning_rate": 0.0002,
+      "loss": 1.4436,
+      "step": 1690
+    },
+    {
+      "epoch": 3.8857142857142857,
+      "grad_norm": 0.7617332935333252,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 3.9085714285714284,
+      "grad_norm": 0.6919480562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9314285714285715,
+      "grad_norm": 0.6987943053245544,
+      "learning_rate": 0.0002,
+      "loss": 1.4317,
+      "step": 1720
+    },
+    {
+      "epoch": 3.954285714285714,
+      "grad_norm": 0.7062228918075562,
+      "learning_rate": 0.0002,
+      "loss": 1.4704,
+      "step": 1730
+    },
+    {
+      "epoch": 3.977142857142857,
+      "grad_norm": 0.6769542098045349,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 1740
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6832144260406494,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 1750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9474865198135376,
+      "eval_runtime": 111.288,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1750
+    },
+    {
+      "epoch": 4.022857142857143,
+      "grad_norm": 1.064110279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 1760
+    },
+    {
+      "epoch": 4.045714285714285,
+      "grad_norm": 0.8380683660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.2013,
+      "step": 1770
+    },
+    {
+      "epoch": 4.0685714285714285,
+      "grad_norm": 1.1863020658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 1780
+    },
+    {
+      "epoch": 4.091428571428572,
+      "grad_norm": 1.0128898620605469,
+      "learning_rate": 0.0002,
+      "loss": 1.2499,
+      "step": 1790
+    },
+    {
+      "epoch": 4.114285714285714,
+      "grad_norm": 0.9221312403678894,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 1800
+    },
+    {
+      "epoch": 4.137142857142857,
+      "grad_norm": 1.1298727989196777,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 1810
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 0.8854547739028931,
+      "learning_rate": 0.0002,
+      "loss": 1.1491,
+      "step": 1820
+    },
+    {
+      "epoch": 4.182857142857143,
+      "grad_norm": 0.8920808434486389,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1830
+    },
+    {
+      "epoch": 4.2057142857142855,
+      "grad_norm": 0.913244366645813,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 1840
+    },
+    {
+      "epoch": 4.228571428571429,
+      "grad_norm": 0.908831000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1850
+    },
+    {
+      "epoch": 4.251428571428572,
+      "grad_norm": 1.0223685503005981,
+      "learning_rate": 0.0002,
+      "loss": 1.1653,
+      "step": 1860
+    },
+    {
+      "epoch": 4.274285714285714,
+      "grad_norm": 0.9771921634674072,
+      "learning_rate": 0.0002,
+      "loss": 1.2497,
+      "step": 1870
+    },
+    {
+      "epoch": 4.297142857142857,
+      "grad_norm": 0.9313384890556335,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 1880
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 1.0754257440567017,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 1890
+    },
+    {
+      "epoch": 4.3428571428571425,
+      "grad_norm": 0.8904672265052795,
+      "learning_rate": 0.0002,
+      "loss": 1.2286,
+      "step": 1900
+    },
+    {
+      "epoch": 4.365714285714286,
+      "grad_norm": 1.046527624130249,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 1910
+    },
+    {
+      "epoch": 4.388571428571429,
+      "grad_norm": 0.9576982855796814,
+      "learning_rate": 0.0002,
+      "loss": 1.2368,
+      "step": 1920
+    },
+    {
+      "epoch": 4.411428571428571,
+      "grad_norm": 0.9278356432914734,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 1930
+    },
+    {
+      "epoch": 4.434285714285714,
+      "grad_norm": 1.1763030290603638,
+      "learning_rate": 0.0002,
+      "loss": 1.2005,
+      "step": 1940
+    },
+    {
+      "epoch": 4.457142857142857,
+      "grad_norm": 0.9183000326156616,
+      "learning_rate": 0.0002,
+      "loss": 1.1541,
+      "step": 1950
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 1.050980806350708,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 1960
+    },
+    {
+      "epoch": 4.502857142857143,
+      "grad_norm": 0.9975392818450928,
+      "learning_rate": 0.0002,
+      "loss": 1.2133,
+      "step": 1970
+    },
+    {
+      "epoch": 4.525714285714286,
+      "grad_norm": 0.990544319152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2312,
+      "step": 1980
+    },
+    {
+      "epoch": 4.548571428571429,
+      "grad_norm": 1.004794955253601,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 1990
+    },
+    {
+      "epoch": 4.571428571428571,
+      "grad_norm": 0.9294857978820801,
+      "learning_rate": 0.0002,
+      "loss": 1.2085,
+      "step": 2000
+    },
+    {
+      "epoch": 4.594285714285714,
+      "grad_norm": 0.93436598777771,
+      "learning_rate": 0.0002,
+      "loss": 1.2874,
+      "step": 2010
+    },
+    {
+      "epoch": 4.617142857142857,
+      "grad_norm": 0.8704655766487122,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 2020
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 0.9077927470207214,
+      "learning_rate": 0.0002,
+      "loss": 1.204,
+      "step": 2030
+    },
+    {
+      "epoch": 4.662857142857143,
+      "grad_norm": 0.912987470626831,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 2040
+    },
+    {
+      "epoch": 4.685714285714286,
+      "grad_norm": 0.9740643501281738,
+      "learning_rate": 0.0002,
+      "loss": 1.2868,
+      "step": 2050
+    },
+    {
+      "epoch": 4.708571428571428,
+      "grad_norm": 1.133357048034668,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 2060
+    },
+    {
+      "epoch": 4.731428571428571,
+      "grad_norm": 0.8844527006149292,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 2070
+    },
+    {
+      "epoch": 4.7542857142857144,
+      "grad_norm": 1.0083311796188354,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 2080
+    },
+    {
+      "epoch": 4.777142857142858,
+      "grad_norm": 1.000447154045105,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 2090
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.9620300531387329,
+      "learning_rate": 0.0002,
+      "loss": 1.2313,
+      "step": 2100
+    },
+    {
+      "epoch": 4.822857142857143,
+      "grad_norm": 0.9843335151672363,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 2110
+    },
+    {
+      "epoch": 4.845714285714286,
+      "grad_norm": 0.9906681180000305,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 2120
+    },
+    {
+      "epoch": 4.868571428571428,
+      "grad_norm": 0.9544073939323425,
+      "learning_rate": 0.0002,
+      "loss": 1.2325,
+      "step": 2130
+    },
+    {
+      "epoch": 4.8914285714285715,
+      "grad_norm": 0.9392994046211243,
+      "learning_rate": 0.0002,
+      "loss": 1.284,
+      "step": 2140
+    },
+    {
+      "epoch": 4.914285714285715,
+      "grad_norm": 1.104519248008728,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2150
+    },
+    {
+      "epoch": 4.937142857142857,
+      "grad_norm": 0.9495956897735596,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 2160
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 0.9696287512779236,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 2170
+    },
+    {
+      "epoch": 4.982857142857143,
+      "grad_norm": 0.9933681488037109,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 2180
+    },
+    {
+      "epoch": 4.998857142857143,
+      "eval_loss": 2.099808692932129,
+      "eval_runtime": 111.2808,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 2187
+    },
+    {
+      "epoch": 5.005714285714285,
+      "grad_norm": 0.9482853412628174,
+      "learning_rate": 0.0002,
+      "loss": 1.2183,
+      "step": 2190
+    },
+    {
+      "epoch": 5.0285714285714285,
+      "grad_norm": 1.6689555644989014,
+      "learning_rate": 0.0002,
+      "loss": 0.9898,
+      "step": 2200
+    },
+    {
+      "epoch": 5.051428571428572,
+      "grad_norm": 1.2019699811935425,
+      "learning_rate": 0.0002,
+      "loss": 0.9741,
+      "step": 2210
+    },
+    {
+      "epoch": 5.074285714285715,
+      "grad_norm": 1.535780429840088,
+      "learning_rate": 0.0002,
+      "loss": 0.9737,
+      "step": 2220
+    },
+    {
+      "epoch": 5.097142857142857,
+      "grad_norm": 1.2061309814453125,
+      "learning_rate": 0.0002,
+      "loss": 0.9494,
+      "step": 2230
+    },
+    {
+      "epoch": 5.12,
+      "grad_norm": 1.1898778676986694,
+      "learning_rate": 0.0002,
+      "loss": 0.9316,
+      "step": 2240
+    },
+    {
+      "epoch": 5.142857142857143,
+      "grad_norm": 1.158898949623108,
+      "learning_rate": 0.0002,
+      "loss": 1.002,
+      "step": 2250
+    },
+    {
+      "epoch": 5.1657142857142855,
+      "grad_norm": 1.370749592781067,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 2260
+    },
+    {
+      "epoch": 5.188571428571429,
+      "grad_norm": 1.314120888710022,
+      "learning_rate": 0.0002,
+      "loss": 0.9365,
+      "step": 2270
+    },
+    {
+      "epoch": 5.211428571428572,
+      "grad_norm": 1.2184966802597046,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 2280
+    },
+    {
+      "epoch": 5.234285714285714,
+      "grad_norm": 1.4833279848098755,
+      "learning_rate": 0.0002,
+      "loss": 0.9407,
+      "step": 2290
+    },
+    {
+      "epoch": 5.257142857142857,
+      "grad_norm": 1.3348219394683838,
+      "learning_rate": 0.0002,
+      "loss": 0.9635,
+      "step": 2300
+    },
+    {
+      "epoch": 5.28,
+      "grad_norm": 1.4166619777679443,
+      "learning_rate": 0.0002,
+      "loss": 1.0294,
+      "step": 2310
+    },
+    {
+      "epoch": 5.3028571428571425,
+      "grad_norm": 1.4539530277252197,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 2320
+    },
+    {
+      "epoch": 5.325714285714286,
+      "grad_norm": 1.4642518758773804,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 2330
+    },
+    {
+      "epoch": 5.348571428571429,
+      "grad_norm": 1.3938848972320557,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 2340
+    },
+    {
+      "epoch": 5.371428571428572,
+      "grad_norm": 1.1147894859313965,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 2350
+    },
+    {
+      "epoch": 5.394285714285714,
+      "grad_norm": 1.3465309143066406,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2360
+    },
+    {
+      "epoch": 5.417142857142857,
+      "grad_norm": 1.4788566827774048,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 2370
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 1.3808705806732178,
+      "learning_rate": 0.0002,
+      "loss": 0.9896,
+      "step": 2380
+    },
+    {
+      "epoch": 5.462857142857143,
+      "grad_norm": 1.2336329221725464,
+      "learning_rate": 0.0002,
+      "loss": 1.0279,
+      "step": 2390
+    },
+    {
+      "epoch": 5.485714285714286,
+      "grad_norm": 1.5445678234100342,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 2400
+    },
+    {
+      "epoch": 5.508571428571429,
+      "grad_norm": 1.107488989830017,
+      "learning_rate": 0.0002,
+      "loss": 0.9534,
+      "step": 2410
+    },
+    {
+      "epoch": 5.531428571428571,
+      "grad_norm": 1.39687979221344,
+      "learning_rate": 0.0002,
+      "loss": 1.0036,
+      "step": 2420
+    },
+    {
+      "epoch": 5.554285714285714,
+      "grad_norm": 1.3905695676803589,
+      "learning_rate": 0.0002,
+      "loss": 0.9959,
+      "step": 2430
+    },
+    {
+      "epoch": 5.577142857142857,
+      "grad_norm": 1.3772821426391602,
+      "learning_rate": 0.0002,
+      "loss": 0.9912,
+      "step": 2440
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 1.1661899089813232,
+      "learning_rate": 0.0002,
+      "loss": 0.9825,
+      "step": 2450
+    },
+    {
+      "epoch": 5.622857142857143,
+      "grad_norm": 1.2730463743209839,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 2460
+    },
+    {
+      "epoch": 5.645714285714286,
+      "grad_norm": 1.2251193523406982,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2470
+    },
+    {
+      "epoch": 5.668571428571429,
+      "grad_norm": 1.5454859733581543,
+      "learning_rate": 0.0002,
+      "loss": 1.079,
+      "step": 2480
+    },
+    {
+      "epoch": 5.691428571428571,
+      "grad_norm": 1.5405735969543457,
+      "learning_rate": 0.0002,
+      "loss": 1.0414,
+      "step": 2490
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.2555434703826904,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 2500
+    },
+    {
+      "epoch": 5.737142857142857,
+      "grad_norm": 1.3323487043380737,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 2510
+    },
+    {
+      "epoch": 5.76,
+      "grad_norm": 1.3106356859207153,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 2520
+    },
+    {
+      "epoch": 5.782857142857143,
+      "grad_norm": 1.4832439422607422,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 2530
+    },
+    {
+      "epoch": 5.805714285714286,
+      "grad_norm": 1.1336562633514404,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 2540
+    },
+    {
+      "epoch": 5.828571428571428,
+      "grad_norm": 1.2434223890304565,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2550
+    },
+    {
+      "epoch": 5.851428571428571,
+      "grad_norm": 1.2825450897216797,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 2560
+    },
+    {
+      "epoch": 5.8742857142857146,
+      "grad_norm": 1.4373180866241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0642,
+      "step": 2570
+    },
+    {
+      "epoch": 5.897142857142857,
+      "grad_norm": 1.435015320777893,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 2580
+    },
+    {
+      "epoch": 5.92,
+      "grad_norm": 1.4075653553009033,
+      "learning_rate": 0.0002,
+      "loss": 1.0272,
+      "step": 2590
+    },
+    {
+      "epoch": 5.942857142857143,
+      "grad_norm": 1.319630742073059,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 2600
+    },
+    {
+      "epoch": 5.965714285714286,
+      "grad_norm": 1.278330683708191,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 2610
+    },
+    {
+      "epoch": 5.988571428571428,
+      "grad_norm": 1.258158564567566,
+      "learning_rate": 0.0002,
+      "loss": 1.0766,
+      "step": 2620
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.3689301013946533,
+      "eval_runtime": 53.9067,
+      "eval_samples_per_second": 9.405,
+      "eval_steps_per_second": 1.187,
+      "step": 2625
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 1.3482983227392e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-2625/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b5957b8c1cf78a037cd4cb28a91e479b2772d34d
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a63a66efe51eade1b9db3a782d99bc39a1e84281614b84989b4348e1f79c779d
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a52a3098e0998ce805cd093bfcf013db26916032
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:966121a4ac71b5d4daac8d4447757f2a8760e183f3705e4ad2e32751089035a5
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..05d9d819b10f6fe669fd791f284b855e1b7e643f
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5d8d725749217e42693110f9b71f15a25465f765c9d4eaa627a8624221589e2
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2036ac3aefac234519f52c8600b40a6386706df9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5da6eb53a5412fee93151405c8188c4782e3828b634238ebc610dcc0daafcddd
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c7080667b59b789e5c80584eb7744ab50cd5aa4
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/trainer_state.json
@@ -0,0 +1,2231 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 6.998857142857143,
+  "eval_steps": 10,
+  "global_step": 3062,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    },
+    {
+      "epoch": 3.0171428571428573,
+      "grad_norm": 0.5184893012046814,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 1320
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.5665355920791626,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0628571428571427,
+      "grad_norm": 0.6601403951644897,
+      "learning_rate": 0.0002,
+      "loss": 1.3741,
+      "step": 1340
+    },
+    {
+      "epoch": 3.085714285714286,
+      "grad_norm": 0.6921621561050415,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 1350
+    },
+    {
+      "epoch": 3.1085714285714285,
+      "grad_norm": 0.6406348943710327,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 1360
+    },
+    {
+      "epoch": 3.1314285714285712,
+      "grad_norm": 0.5814554691314697,
+      "learning_rate": 0.0002,
+      "loss": 1.3563,
+      "step": 1370
+    },
+    {
+      "epoch": 3.1542857142857144,
+      "grad_norm": 0.683325469493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 1380
+    },
+    {
+      "epoch": 3.177142857142857,
+      "grad_norm": 0.6686155200004578,
+      "learning_rate": 0.0002,
+      "loss": 1.4106,
+      "step": 1390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8159713745117188,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 1400
+    },
+    {
+      "epoch": 3.222857142857143,
+      "grad_norm": 0.646216094493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4279,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2457142857142856,
+      "grad_norm": 0.7323529720306396,
+      "learning_rate": 0.0002,
+      "loss": 1.4232,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2685714285714287,
+      "grad_norm": 0.689349353313446,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2914285714285714,
+      "grad_norm": 0.727894127368927,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 1440
+    },
+    {
+      "epoch": 3.314285714285714,
+      "grad_norm": 0.6921590566635132,
+      "learning_rate": 0.0002,
+      "loss": 1.4,
+      "step": 1450
+    },
+    {
+      "epoch": 3.337142857142857,
+      "grad_norm": 0.6176243424415588,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 1460
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.9006354212760925,
+      "learning_rate": 0.0002,
+      "loss": 1.4323,
+      "step": 1470
+    },
+    {
+      "epoch": 3.382857142857143,
+      "grad_norm": 0.8145929574966431,
+      "learning_rate": 0.0002,
+      "loss": 1.4353,
+      "step": 1480
+    },
+    {
+      "epoch": 3.4057142857142857,
+      "grad_norm": 0.6640016436576843,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.7266780138015747,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4514285714285715,
+      "grad_norm": 0.9351356029510498,
+      "learning_rate": 0.0002,
+      "loss": 1.4108,
+      "step": 1510
+    },
+    {
+      "epoch": 3.474285714285714,
+      "grad_norm": 0.675645649433136,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4971428571428573,
+      "grad_norm": 0.761472225189209,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 1530
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.6653069257736206,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5428571428571427,
+      "grad_norm": 0.667412519454956,
+      "learning_rate": 0.0002,
+      "loss": 1.4686,
+      "step": 1550
+    },
+    {
+      "epoch": 3.565714285714286,
+      "grad_norm": 0.6395593881607056,
+      "learning_rate": 0.0002,
+      "loss": 1.4241,
+      "step": 1560
+    },
+    {
+      "epoch": 3.5885714285714285,
+      "grad_norm": 0.7588621377944946,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1570
+    },
+    {
+      "epoch": 3.611428571428571,
+      "grad_norm": 0.6206456422805786,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6342857142857143,
+      "grad_norm": 0.7591291666030884,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 1590
+    },
+    {
+      "epoch": 3.657142857142857,
+      "grad_norm": 0.6476313471794128,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 1600
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.6731392741203308,
+      "learning_rate": 0.0002,
+      "loss": 1.4598,
+      "step": 1610
+    },
+    {
+      "epoch": 3.702857142857143,
+      "grad_norm": 0.725190281867981,
+      "learning_rate": 0.0002,
+      "loss": 1.4225,
+      "step": 1620
+    },
+    {
+      "epoch": 3.725714285714286,
+      "grad_norm": 0.6720049977302551,
+      "learning_rate": 0.0002,
+      "loss": 1.4525,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7485714285714287,
+      "grad_norm": 0.6301007270812988,
+      "learning_rate": 0.0002,
+      "loss": 1.429,
+      "step": 1640
+    },
+    {
+      "epoch": 3.7714285714285714,
+      "grad_norm": 0.715893566608429,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 1650
+    },
+    {
+      "epoch": 3.7942857142857145,
+      "grad_norm": 0.7539359927177429,
+      "learning_rate": 0.0002,
+      "loss": 1.3624,
+      "step": 1660
+    },
+    {
+      "epoch": 3.817142857142857,
+      "grad_norm": 0.6658543348312378,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 1670
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.7019526958465576,
+      "learning_rate": 0.0002,
+      "loss": 1.3934,
+      "step": 1680
+    },
+    {
+      "epoch": 3.862857142857143,
+      "grad_norm": 0.6517802476882935,
+      "learning_rate": 0.0002,
+      "loss": 1.4436,
+      "step": 1690
+    },
+    {
+      "epoch": 3.8857142857142857,
+      "grad_norm": 0.7617332935333252,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 3.9085714285714284,
+      "grad_norm": 0.6919480562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9314285714285715,
+      "grad_norm": 0.6987943053245544,
+      "learning_rate": 0.0002,
+      "loss": 1.4317,
+      "step": 1720
+    },
+    {
+      "epoch": 3.954285714285714,
+      "grad_norm": 0.7062228918075562,
+      "learning_rate": 0.0002,
+      "loss": 1.4704,
+      "step": 1730
+    },
+    {
+      "epoch": 3.977142857142857,
+      "grad_norm": 0.6769542098045349,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 1740
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6832144260406494,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 1750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9474865198135376,
+      "eval_runtime": 111.288,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1750
+    },
+    {
+      "epoch": 4.022857142857143,
+      "grad_norm": 1.064110279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 1760
+    },
+    {
+      "epoch": 4.045714285714285,
+      "grad_norm": 0.8380683660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.2013,
+      "step": 1770
+    },
+    {
+      "epoch": 4.0685714285714285,
+      "grad_norm": 1.1863020658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 1780
+    },
+    {
+      "epoch": 4.091428571428572,
+      "grad_norm": 1.0128898620605469,
+      "learning_rate": 0.0002,
+      "loss": 1.2499,
+      "step": 1790
+    },
+    {
+      "epoch": 4.114285714285714,
+      "grad_norm": 0.9221312403678894,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 1800
+    },
+    {
+      "epoch": 4.137142857142857,
+      "grad_norm": 1.1298727989196777,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 1810
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 0.8854547739028931,
+      "learning_rate": 0.0002,
+      "loss": 1.1491,
+      "step": 1820
+    },
+    {
+      "epoch": 4.182857142857143,
+      "grad_norm": 0.8920808434486389,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1830
+    },
+    {
+      "epoch": 4.2057142857142855,
+      "grad_norm": 0.913244366645813,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 1840
+    },
+    {
+      "epoch": 4.228571428571429,
+      "grad_norm": 0.908831000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1850
+    },
+    {
+      "epoch": 4.251428571428572,
+      "grad_norm": 1.0223685503005981,
+      "learning_rate": 0.0002,
+      "loss": 1.1653,
+      "step": 1860
+    },
+    {
+      "epoch": 4.274285714285714,
+      "grad_norm": 0.9771921634674072,
+      "learning_rate": 0.0002,
+      "loss": 1.2497,
+      "step": 1870
+    },
+    {
+      "epoch": 4.297142857142857,
+      "grad_norm": 0.9313384890556335,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 1880
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 1.0754257440567017,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 1890
+    },
+    {
+      "epoch": 4.3428571428571425,
+      "grad_norm": 0.8904672265052795,
+      "learning_rate": 0.0002,
+      "loss": 1.2286,
+      "step": 1900
+    },
+    {
+      "epoch": 4.365714285714286,
+      "grad_norm": 1.046527624130249,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 1910
+    },
+    {
+      "epoch": 4.388571428571429,
+      "grad_norm": 0.9576982855796814,
+      "learning_rate": 0.0002,
+      "loss": 1.2368,
+      "step": 1920
+    },
+    {
+      "epoch": 4.411428571428571,
+      "grad_norm": 0.9278356432914734,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 1930
+    },
+    {
+      "epoch": 4.434285714285714,
+      "grad_norm": 1.1763030290603638,
+      "learning_rate": 0.0002,
+      "loss": 1.2005,
+      "step": 1940
+    },
+    {
+      "epoch": 4.457142857142857,
+      "grad_norm": 0.9183000326156616,
+      "learning_rate": 0.0002,
+      "loss": 1.1541,
+      "step": 1950
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 1.050980806350708,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 1960
+    },
+    {
+      "epoch": 4.502857142857143,
+      "grad_norm": 0.9975392818450928,
+      "learning_rate": 0.0002,
+      "loss": 1.2133,
+      "step": 1970
+    },
+    {
+      "epoch": 4.525714285714286,
+      "grad_norm": 0.990544319152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2312,
+      "step": 1980
+    },
+    {
+      "epoch": 4.548571428571429,
+      "grad_norm": 1.004794955253601,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 1990
+    },
+    {
+      "epoch": 4.571428571428571,
+      "grad_norm": 0.9294857978820801,
+      "learning_rate": 0.0002,
+      "loss": 1.2085,
+      "step": 2000
+    },
+    {
+      "epoch": 4.594285714285714,
+      "grad_norm": 0.93436598777771,
+      "learning_rate": 0.0002,
+      "loss": 1.2874,
+      "step": 2010
+    },
+    {
+      "epoch": 4.617142857142857,
+      "grad_norm": 0.8704655766487122,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 2020
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 0.9077927470207214,
+      "learning_rate": 0.0002,
+      "loss": 1.204,
+      "step": 2030
+    },
+    {
+      "epoch": 4.662857142857143,
+      "grad_norm": 0.912987470626831,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 2040
+    },
+    {
+      "epoch": 4.685714285714286,
+      "grad_norm": 0.9740643501281738,
+      "learning_rate": 0.0002,
+      "loss": 1.2868,
+      "step": 2050
+    },
+    {
+      "epoch": 4.708571428571428,
+      "grad_norm": 1.133357048034668,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 2060
+    },
+    {
+      "epoch": 4.731428571428571,
+      "grad_norm": 0.8844527006149292,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 2070
+    },
+    {
+      "epoch": 4.7542857142857144,
+      "grad_norm": 1.0083311796188354,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 2080
+    },
+    {
+      "epoch": 4.777142857142858,
+      "grad_norm": 1.000447154045105,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 2090
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.9620300531387329,
+      "learning_rate": 0.0002,
+      "loss": 1.2313,
+      "step": 2100
+    },
+    {
+      "epoch": 4.822857142857143,
+      "grad_norm": 0.9843335151672363,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 2110
+    },
+    {
+      "epoch": 4.845714285714286,
+      "grad_norm": 0.9906681180000305,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 2120
+    },
+    {
+      "epoch": 4.868571428571428,
+      "grad_norm": 0.9544073939323425,
+      "learning_rate": 0.0002,
+      "loss": 1.2325,
+      "step": 2130
+    },
+    {
+      "epoch": 4.8914285714285715,
+      "grad_norm": 0.9392994046211243,
+      "learning_rate": 0.0002,
+      "loss": 1.284,
+      "step": 2140
+    },
+    {
+      "epoch": 4.914285714285715,
+      "grad_norm": 1.104519248008728,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2150
+    },
+    {
+      "epoch": 4.937142857142857,
+      "grad_norm": 0.9495956897735596,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 2160
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 0.9696287512779236,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 2170
+    },
+    {
+      "epoch": 4.982857142857143,
+      "grad_norm": 0.9933681488037109,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 2180
+    },
+    {
+      "epoch": 4.998857142857143,
+      "eval_loss": 2.099808692932129,
+      "eval_runtime": 111.2808,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 2187
+    },
+    {
+      "epoch": 5.005714285714285,
+      "grad_norm": 0.9482853412628174,
+      "learning_rate": 0.0002,
+      "loss": 1.2183,
+      "step": 2190
+    },
+    {
+      "epoch": 5.0285714285714285,
+      "grad_norm": 1.6689555644989014,
+      "learning_rate": 0.0002,
+      "loss": 0.9898,
+      "step": 2200
+    },
+    {
+      "epoch": 5.051428571428572,
+      "grad_norm": 1.2019699811935425,
+      "learning_rate": 0.0002,
+      "loss": 0.9741,
+      "step": 2210
+    },
+    {
+      "epoch": 5.074285714285715,
+      "grad_norm": 1.535780429840088,
+      "learning_rate": 0.0002,
+      "loss": 0.9737,
+      "step": 2220
+    },
+    {
+      "epoch": 5.097142857142857,
+      "grad_norm": 1.2061309814453125,
+      "learning_rate": 0.0002,
+      "loss": 0.9494,
+      "step": 2230
+    },
+    {
+      "epoch": 5.12,
+      "grad_norm": 1.1898778676986694,
+      "learning_rate": 0.0002,
+      "loss": 0.9316,
+      "step": 2240
+    },
+    {
+      "epoch": 5.142857142857143,
+      "grad_norm": 1.158898949623108,
+      "learning_rate": 0.0002,
+      "loss": 1.002,
+      "step": 2250
+    },
+    {
+      "epoch": 5.1657142857142855,
+      "grad_norm": 1.370749592781067,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 2260
+    },
+    {
+      "epoch": 5.188571428571429,
+      "grad_norm": 1.314120888710022,
+      "learning_rate": 0.0002,
+      "loss": 0.9365,
+      "step": 2270
+    },
+    {
+      "epoch": 5.211428571428572,
+      "grad_norm": 1.2184966802597046,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 2280
+    },
+    {
+      "epoch": 5.234285714285714,
+      "grad_norm": 1.4833279848098755,
+      "learning_rate": 0.0002,
+      "loss": 0.9407,
+      "step": 2290
+    },
+    {
+      "epoch": 5.257142857142857,
+      "grad_norm": 1.3348219394683838,
+      "learning_rate": 0.0002,
+      "loss": 0.9635,
+      "step": 2300
+    },
+    {
+      "epoch": 5.28,
+      "grad_norm": 1.4166619777679443,
+      "learning_rate": 0.0002,
+      "loss": 1.0294,
+      "step": 2310
+    },
+    {
+      "epoch": 5.3028571428571425,
+      "grad_norm": 1.4539530277252197,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 2320
+    },
+    {
+      "epoch": 5.325714285714286,
+      "grad_norm": 1.4642518758773804,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 2330
+    },
+    {
+      "epoch": 5.348571428571429,
+      "grad_norm": 1.3938848972320557,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 2340
+    },
+    {
+      "epoch": 5.371428571428572,
+      "grad_norm": 1.1147894859313965,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 2350
+    },
+    {
+      "epoch": 5.394285714285714,
+      "grad_norm": 1.3465309143066406,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2360
+    },
+    {
+      "epoch": 5.417142857142857,
+      "grad_norm": 1.4788566827774048,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 2370
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 1.3808705806732178,
+      "learning_rate": 0.0002,
+      "loss": 0.9896,
+      "step": 2380
+    },
+    {
+      "epoch": 5.462857142857143,
+      "grad_norm": 1.2336329221725464,
+      "learning_rate": 0.0002,
+      "loss": 1.0279,
+      "step": 2390
+    },
+    {
+      "epoch": 5.485714285714286,
+      "grad_norm": 1.5445678234100342,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 2400
+    },
+    {
+      "epoch": 5.508571428571429,
+      "grad_norm": 1.107488989830017,
+      "learning_rate": 0.0002,
+      "loss": 0.9534,
+      "step": 2410
+    },
+    {
+      "epoch": 5.531428571428571,
+      "grad_norm": 1.39687979221344,
+      "learning_rate": 0.0002,
+      "loss": 1.0036,
+      "step": 2420
+    },
+    {
+      "epoch": 5.554285714285714,
+      "grad_norm": 1.3905695676803589,
+      "learning_rate": 0.0002,
+      "loss": 0.9959,
+      "step": 2430
+    },
+    {
+      "epoch": 5.577142857142857,
+      "grad_norm": 1.3772821426391602,
+      "learning_rate": 0.0002,
+      "loss": 0.9912,
+      "step": 2440
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 1.1661899089813232,
+      "learning_rate": 0.0002,
+      "loss": 0.9825,
+      "step": 2450
+    },
+    {
+      "epoch": 5.622857142857143,
+      "grad_norm": 1.2730463743209839,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 2460
+    },
+    {
+      "epoch": 5.645714285714286,
+      "grad_norm": 1.2251193523406982,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2470
+    },
+    {
+      "epoch": 5.668571428571429,
+      "grad_norm": 1.5454859733581543,
+      "learning_rate": 0.0002,
+      "loss": 1.079,
+      "step": 2480
+    },
+    {
+      "epoch": 5.691428571428571,
+      "grad_norm": 1.5405735969543457,
+      "learning_rate": 0.0002,
+      "loss": 1.0414,
+      "step": 2490
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.2555434703826904,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 2500
+    },
+    {
+      "epoch": 5.737142857142857,
+      "grad_norm": 1.3323487043380737,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 2510
+    },
+    {
+      "epoch": 5.76,
+      "grad_norm": 1.3106356859207153,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 2520
+    },
+    {
+      "epoch": 5.782857142857143,
+      "grad_norm": 1.4832439422607422,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 2530
+    },
+    {
+      "epoch": 5.805714285714286,
+      "grad_norm": 1.1336562633514404,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 2540
+    },
+    {
+      "epoch": 5.828571428571428,
+      "grad_norm": 1.2434223890304565,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2550
+    },
+    {
+      "epoch": 5.851428571428571,
+      "grad_norm": 1.2825450897216797,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 2560
+    },
+    {
+      "epoch": 5.8742857142857146,
+      "grad_norm": 1.4373180866241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0642,
+      "step": 2570
+    },
+    {
+      "epoch": 5.897142857142857,
+      "grad_norm": 1.435015320777893,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 2580
+    },
+    {
+      "epoch": 5.92,
+      "grad_norm": 1.4075653553009033,
+      "learning_rate": 0.0002,
+      "loss": 1.0272,
+      "step": 2590
+    },
+    {
+      "epoch": 5.942857142857143,
+      "grad_norm": 1.319630742073059,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 2600
+    },
+    {
+      "epoch": 5.965714285714286,
+      "grad_norm": 1.278330683708191,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 2610
+    },
+    {
+      "epoch": 5.988571428571428,
+      "grad_norm": 1.258158564567566,
+      "learning_rate": 0.0002,
+      "loss": 1.0766,
+      "step": 2620
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.3689301013946533,
+      "eval_runtime": 53.9067,
+      "eval_samples_per_second": 9.405,
+      "eval_steps_per_second": 1.187,
+      "step": 2625
+    },
+    {
+      "epoch": 6.011428571428572,
+      "grad_norm": 1.3128368854522705,
+      "learning_rate": 0.0002,
+      "loss": 0.9142,
+      "step": 2630
+    },
+    {
+      "epoch": 6.034285714285715,
+      "grad_norm": 1.4280474185943604,
+      "learning_rate": 0.0002,
+      "loss": 0.7716,
+      "step": 2640
+    },
+    {
+      "epoch": 6.057142857142857,
+      "grad_norm": 1.5061450004577637,
+      "learning_rate": 0.0002,
+      "loss": 0.7776,
+      "step": 2650
+    },
+    {
+      "epoch": 6.08,
+      "grad_norm": 1.6013342142105103,
+      "learning_rate": 0.0002,
+      "loss": 0.7707,
+      "step": 2660
+    },
+    {
+      "epoch": 6.102857142857143,
+      "grad_norm": 2.0107381343841553,
+      "learning_rate": 0.0002,
+      "loss": 0.7543,
+      "step": 2670
+    },
+    {
+      "epoch": 6.1257142857142854,
+      "grad_norm": 1.5010124444961548,
+      "learning_rate": 0.0002,
+      "loss": 0.747,
+      "step": 2680
+    },
+    {
+      "epoch": 6.148571428571429,
+      "grad_norm": 1.5222150087356567,
+      "learning_rate": 0.0002,
+      "loss": 0.7501,
+      "step": 2690
+    },
+    {
+      "epoch": 6.171428571428572,
+      "grad_norm": 1.5413103103637695,
+      "learning_rate": 0.0002,
+      "loss": 0.7712,
+      "step": 2700
+    },
+    {
+      "epoch": 6.194285714285714,
+      "grad_norm": 1.527140736579895,
+      "learning_rate": 0.0002,
+      "loss": 0.7,
+      "step": 2710
+    },
+    {
+      "epoch": 6.217142857142857,
+      "grad_norm": 1.9386590719223022,
+      "learning_rate": 0.0002,
+      "loss": 0.7539,
+      "step": 2720
+    },
+    {
+      "epoch": 6.24,
+      "grad_norm": 1.8115214109420776,
+      "learning_rate": 0.0002,
+      "loss": 0.7586,
+      "step": 2730
+    },
+    {
+      "epoch": 6.2628571428571425,
+      "grad_norm": 1.6221802234649658,
+      "learning_rate": 0.0002,
+      "loss": 0.7426,
+      "step": 2740
+    },
+    {
+      "epoch": 6.285714285714286,
+      "grad_norm": 1.6698768138885498,
+      "learning_rate": 0.0002,
+      "loss": 0.8002,
+      "step": 2750
+    },
+    {
+      "epoch": 6.308571428571429,
+      "grad_norm": 1.7960610389709473,
+      "learning_rate": 0.0002,
+      "loss": 0.7293,
+      "step": 2760
+    },
+    {
+      "epoch": 6.331428571428571,
+      "grad_norm": 1.32172429561615,
+      "learning_rate": 0.0002,
+      "loss": 0.7405,
+      "step": 2770
+    },
+    {
+      "epoch": 6.354285714285714,
+      "grad_norm": 1.7468090057373047,
+      "learning_rate": 0.0002,
+      "loss": 0.7198,
+      "step": 2780
+    },
+    {
+      "epoch": 6.377142857142857,
+      "grad_norm": 1.6777397394180298,
+      "learning_rate": 0.0002,
+      "loss": 0.76,
+      "step": 2790
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 1.6200671195983887,
+      "learning_rate": 0.0002,
+      "loss": 0.7879,
+      "step": 2800
+    },
+    {
+      "epoch": 6.422857142857143,
+      "grad_norm": 1.723505973815918,
+      "learning_rate": 0.0002,
+      "loss": 0.7807,
+      "step": 2810
+    },
+    {
+      "epoch": 6.445714285714286,
+      "grad_norm": 1.4945589303970337,
+      "learning_rate": 0.0002,
+      "loss": 0.8645,
+      "step": 2820
+    },
+    {
+      "epoch": 6.468571428571429,
+      "grad_norm": 1.666458010673523,
+      "learning_rate": 0.0002,
+      "loss": 0.809,
+      "step": 2830
+    },
+    {
+      "epoch": 6.491428571428571,
+      "grad_norm": 1.6586525440216064,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 2840
+    },
+    {
+      "epoch": 6.514285714285714,
+      "grad_norm": 1.7480043172836304,
+      "learning_rate": 0.0002,
+      "loss": 0.8062,
+      "step": 2850
+    },
+    {
+      "epoch": 6.537142857142857,
+      "grad_norm": 1.4605649709701538,
+      "learning_rate": 0.0002,
+      "loss": 0.7602,
+      "step": 2860
+    },
+    {
+      "epoch": 6.5600000000000005,
+      "grad_norm": 1.4841814041137695,
+      "learning_rate": 0.0002,
+      "loss": 0.8186,
+      "step": 2870
+    },
+    {
+      "epoch": 6.582857142857143,
+      "grad_norm": 1.4653114080429077,
+      "learning_rate": 0.0002,
+      "loss": 0.8156,
+      "step": 2880
+    },
+    {
+      "epoch": 6.605714285714286,
+      "grad_norm": 1.7266837358474731,
+      "learning_rate": 0.0002,
+      "loss": 0.8111,
+      "step": 2890
+    },
+    {
+      "epoch": 6.628571428571428,
+      "grad_norm": 1.4860098361968994,
+      "learning_rate": 0.0002,
+      "loss": 0.7644,
+      "step": 2900
+    },
+    {
+      "epoch": 6.651428571428571,
+      "grad_norm": 1.7177597284317017,
+      "learning_rate": 0.0002,
+      "loss": 0.7991,
+      "step": 2910
+    },
+    {
+      "epoch": 6.674285714285714,
+      "grad_norm": 1.6757104396820068,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 2920
+    },
+    {
+      "epoch": 6.6971428571428575,
+      "grad_norm": 1.5177433490753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 2930
+    },
+    {
+      "epoch": 6.72,
+      "grad_norm": 1.8073889017105103,
+      "learning_rate": 0.0002,
+      "loss": 0.7825,
+      "step": 2940
+    },
+    {
+      "epoch": 6.742857142857143,
+      "grad_norm": 1.72337007522583,
+      "learning_rate": 0.0002,
+      "loss": 0.8234,
+      "step": 2950
+    },
+    {
+      "epoch": 6.765714285714286,
+      "grad_norm": 1.6298240423202515,
+      "learning_rate": 0.0002,
+      "loss": 0.896,
+      "step": 2960
+    },
+    {
+      "epoch": 6.788571428571428,
+      "grad_norm": 1.6140344142913818,
+      "learning_rate": 0.0002,
+      "loss": 0.8252,
+      "step": 2970
+    },
+    {
+      "epoch": 6.811428571428571,
+      "grad_norm": 1.7180862426757812,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 2980
+    },
+    {
+      "epoch": 6.8342857142857145,
+      "grad_norm": 1.7589894533157349,
+      "learning_rate": 0.0002,
+      "loss": 0.7929,
+      "step": 2990
+    },
+    {
+      "epoch": 6.857142857142857,
+      "grad_norm": 1.780195713043213,
+      "learning_rate": 0.0002,
+      "loss": 0.828,
+      "step": 3000
+    },
+    {
+      "epoch": 6.88,
+      "grad_norm": 1.7182508707046509,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 3010
+    },
+    {
+      "epoch": 6.902857142857143,
+      "grad_norm": 1.6308406591415405,
+      "learning_rate": 0.0002,
+      "loss": 0.7964,
+      "step": 3020
+    },
+    {
+      "epoch": 6.925714285714285,
+      "grad_norm": 1.5080229043960571,
+      "learning_rate": 0.0002,
+      "loss": 0.8207,
+      "step": 3030
+    },
+    {
+      "epoch": 6.948571428571428,
+      "grad_norm": 1.623555064201355,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 3040
+    },
+    {
+      "epoch": 6.9714285714285715,
+      "grad_norm": 1.526054859161377,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 3050
+    },
+    {
+      "epoch": 6.994285714285715,
+      "grad_norm": 1.6671174764633179,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3060
+    },
+    {
+      "epoch": 6.998857142857143,
+      "eval_loss": 2.647613525390625,
+      "eval_runtime": 111.2255,
+      "eval_samples_per_second": 4.558,
+      "eval_steps_per_second": 0.575,
+      "step": 3062
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 1.5730147098624e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3062/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2900063d6dcf382201c28a35fb26eec7d38239cb
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52c050b52221caafacf459aff714f0d1f0e747a1dcb029bc27d138d6ae22ecdb
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ee345e616e86721f680d98615f1d98a76693bab1
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4a5d2bf5be3b702f111f7a0a3723431a90f8b3eb84348e2c27080a9d62a5c32
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e289d2ab4f0910ba83fcff7962458d8e08d8b89c
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:185518ed415551a71a8c312313b99c2e0eed937898153ee7cae6d3b9d512de56
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b12c82193244f9aef9e538eef6ec5730bf8a0a88
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a6cdc16cfd9c080cdc06d33762e71556a697b7c6f820509ad58b752ec1a3b6
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..d4a70e57c49993e56457f96bbd3bcf48434c2d55
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/trainer_state.json
@@ -0,0 +1,2540 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 7.990857142857143,
+  "eval_steps": 10,
+  "global_step": 3496,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    },
+    {
+      "epoch": 2.0114285714285716,
+      "grad_norm": 0.33838793635368347,
+      "learning_rate": 0.0002,
+      "loss": 1.7215,
+      "step": 880
+    },
+    {
+      "epoch": 2.0342857142857143,
+      "grad_norm": 0.4219334125518799,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 890
+    },
+    {
+      "epoch": 2.057142857142857,
+      "grad_norm": 0.43962377309799194,
+      "learning_rate": 0.0002,
+      "loss": 1.6144,
+      "step": 900
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 0.41956576704978943,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 910
+    },
+    {
+      "epoch": 2.1028571428571428,
+      "grad_norm": 0.4439629912376404,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 920
+    },
+    {
+      "epoch": 2.125714285714286,
+      "grad_norm": 0.43405696749687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 930
+    },
+    {
+      "epoch": 2.1485714285714286,
+      "grad_norm": 0.4321737587451935,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 940
+    },
+    {
+      "epoch": 2.1714285714285713,
+      "grad_norm": 0.4689100682735443,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 950
+    },
+    {
+      "epoch": 2.1942857142857144,
+      "grad_norm": 0.47024697065353394,
+      "learning_rate": 0.0002,
+      "loss": 1.6442,
+      "step": 960
+    },
+    {
+      "epoch": 2.217142857142857,
+      "grad_norm": 0.4535103440284729,
+      "learning_rate": 0.0002,
+      "loss": 1.4703,
+      "step": 970
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.45990121364593506,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 980
+    },
+    {
+      "epoch": 2.262857142857143,
+      "grad_norm": 0.48427215218544006,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 990
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.43076643347740173,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3085714285714287,
+      "grad_norm": 0.4854483902454376,
+      "learning_rate": 0.0002,
+      "loss": 1.5938,
+      "step": 1010
+    },
+    {
+      "epoch": 2.3314285714285714,
+      "grad_norm": 0.46086496114730835,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 1020
+    },
+    {
+      "epoch": 2.354285714285714,
+      "grad_norm": 0.4714847505092621,
+      "learning_rate": 0.0002,
+      "loss": 1.6392,
+      "step": 1030
+    },
+    {
+      "epoch": 2.3771428571428572,
+      "grad_norm": 0.4423409402370453,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 1040
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.46261295676231384,
+      "learning_rate": 0.0002,
+      "loss": 1.5821,
+      "step": 1050
+    },
+    {
+      "epoch": 2.422857142857143,
+      "grad_norm": 0.4914337396621704,
+      "learning_rate": 0.0002,
+      "loss": 1.5503,
+      "step": 1060
+    },
+    {
+      "epoch": 2.4457142857142857,
+      "grad_norm": 0.45144036412239075,
+      "learning_rate": 0.0002,
+      "loss": 1.6459,
+      "step": 1070
+    },
+    {
+      "epoch": 2.4685714285714284,
+      "grad_norm": 0.4510825276374817,
+      "learning_rate": 0.0002,
+      "loss": 1.6416,
+      "step": 1080
+    },
+    {
+      "epoch": 2.4914285714285715,
+      "grad_norm": 0.48552489280700684,
+      "learning_rate": 0.0002,
+      "loss": 1.5808,
+      "step": 1090
+    },
+    {
+      "epoch": 2.5142857142857142,
+      "grad_norm": 0.4768163859844208,
+      "learning_rate": 0.0002,
+      "loss": 1.6659,
+      "step": 1100
+    },
+    {
+      "epoch": 2.5371428571428574,
+      "grad_norm": 0.5192609429359436,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 1110
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 0.49308598041534424,
+      "learning_rate": 0.0002,
+      "loss": 1.5581,
+      "step": 1120
+    },
+    {
+      "epoch": 2.5828571428571427,
+      "grad_norm": 0.5068584084510803,
+      "learning_rate": 0.0002,
+      "loss": 1.6767,
+      "step": 1130
+    },
+    {
+      "epoch": 2.605714285714286,
+      "grad_norm": 0.4822661280632019,
+      "learning_rate": 0.0002,
+      "loss": 1.5788,
+      "step": 1140
+    },
+    {
+      "epoch": 2.6285714285714286,
+      "grad_norm": 0.5028144717216492,
+      "learning_rate": 0.0002,
+      "loss": 1.6891,
+      "step": 1150
+    },
+    {
+      "epoch": 2.6514285714285712,
+      "grad_norm": 0.48315200209617615,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 1160
+    },
+    {
+      "epoch": 2.6742857142857144,
+      "grad_norm": 0.551934540271759,
+      "learning_rate": 0.0002,
+      "loss": 1.5508,
+      "step": 1170
+    },
+    {
+      "epoch": 2.697142857142857,
+      "grad_norm": 0.49223729968070984,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 1180
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 0.514847457408905,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 1190
+    },
+    {
+      "epoch": 2.742857142857143,
+      "grad_norm": 0.4830605387687683,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 1200
+    },
+    {
+      "epoch": 2.7657142857142856,
+      "grad_norm": 0.4584822952747345,
+      "learning_rate": 0.0002,
+      "loss": 1.6285,
+      "step": 1210
+    },
+    {
+      "epoch": 2.7885714285714287,
+      "grad_norm": 0.4688762426376343,
+      "learning_rate": 0.0002,
+      "loss": 1.6043,
+      "step": 1220
+    },
+    {
+      "epoch": 2.8114285714285714,
+      "grad_norm": 0.4488156735897064,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1230
+    },
+    {
+      "epoch": 2.8342857142857145,
+      "grad_norm": 0.4700278639793396,
+      "learning_rate": 0.0002,
+      "loss": 1.5523,
+      "step": 1240
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.5282207131385803,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 1250
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 0.4874219000339508,
+      "learning_rate": 0.0002,
+      "loss": 1.5532,
+      "step": 1260
+    },
+    {
+      "epoch": 2.902857142857143,
+      "grad_norm": 0.49468332529067993,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 1270
+    },
+    {
+      "epoch": 2.9257142857142857,
+      "grad_norm": 0.49770233035087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 1280
+    },
+    {
+      "epoch": 2.9485714285714284,
+      "grad_norm": 0.4433252811431885,
+      "learning_rate": 0.0002,
+      "loss": 1.606,
+      "step": 1290
+    },
+    {
+      "epoch": 2.9714285714285715,
+      "grad_norm": 0.46836379170417786,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 1300
+    },
+    {
+      "epoch": 2.994285714285714,
+      "grad_norm": 0.5001904368400574,
+      "learning_rate": 0.0002,
+      "loss": 1.5555,
+      "step": 1310
+    },
+    {
+      "epoch": 2.998857142857143,
+      "eval_loss": 1.8684407472610474,
+      "eval_runtime": 111.2835,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1312
+    },
+    {
+      "epoch": 3.0171428571428573,
+      "grad_norm": 0.5184893012046814,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 1320
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.5665355920791626,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 1330
+    },
+    {
+      "epoch": 3.0628571428571427,
+      "grad_norm": 0.6601403951644897,
+      "learning_rate": 0.0002,
+      "loss": 1.3741,
+      "step": 1340
+    },
+    {
+      "epoch": 3.085714285714286,
+      "grad_norm": 0.6921621561050415,
+      "learning_rate": 0.0002,
+      "loss": 1.433,
+      "step": 1350
+    },
+    {
+      "epoch": 3.1085714285714285,
+      "grad_norm": 0.6406348943710327,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 1360
+    },
+    {
+      "epoch": 3.1314285714285712,
+      "grad_norm": 0.5814554691314697,
+      "learning_rate": 0.0002,
+      "loss": 1.3563,
+      "step": 1370
+    },
+    {
+      "epoch": 3.1542857142857144,
+      "grad_norm": 0.683325469493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4096,
+      "step": 1380
+    },
+    {
+      "epoch": 3.177142857142857,
+      "grad_norm": 0.6686155200004578,
+      "learning_rate": 0.0002,
+      "loss": 1.4106,
+      "step": 1390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8159713745117188,
+      "learning_rate": 0.0002,
+      "loss": 1.4394,
+      "step": 1400
+    },
+    {
+      "epoch": 3.222857142857143,
+      "grad_norm": 0.646216094493866,
+      "learning_rate": 0.0002,
+      "loss": 1.4279,
+      "step": 1410
+    },
+    {
+      "epoch": 3.2457142857142856,
+      "grad_norm": 0.7323529720306396,
+      "learning_rate": 0.0002,
+      "loss": 1.4232,
+      "step": 1420
+    },
+    {
+      "epoch": 3.2685714285714287,
+      "grad_norm": 0.689349353313446,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1430
+    },
+    {
+      "epoch": 3.2914285714285714,
+      "grad_norm": 0.727894127368927,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 1440
+    },
+    {
+      "epoch": 3.314285714285714,
+      "grad_norm": 0.6921590566635132,
+      "learning_rate": 0.0002,
+      "loss": 1.4,
+      "step": 1450
+    },
+    {
+      "epoch": 3.337142857142857,
+      "grad_norm": 0.6176243424415588,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 1460
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.9006354212760925,
+      "learning_rate": 0.0002,
+      "loss": 1.4323,
+      "step": 1470
+    },
+    {
+      "epoch": 3.382857142857143,
+      "grad_norm": 0.8145929574966431,
+      "learning_rate": 0.0002,
+      "loss": 1.4353,
+      "step": 1480
+    },
+    {
+      "epoch": 3.4057142857142857,
+      "grad_norm": 0.6640016436576843,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 1490
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.7266780138015747,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 1500
+    },
+    {
+      "epoch": 3.4514285714285715,
+      "grad_norm": 0.9351356029510498,
+      "learning_rate": 0.0002,
+      "loss": 1.4108,
+      "step": 1510
+    },
+    {
+      "epoch": 3.474285714285714,
+      "grad_norm": 0.675645649433136,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 1520
+    },
+    {
+      "epoch": 3.4971428571428573,
+      "grad_norm": 0.761472225189209,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 1530
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 0.6653069257736206,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1540
+    },
+    {
+      "epoch": 3.5428571428571427,
+      "grad_norm": 0.667412519454956,
+      "learning_rate": 0.0002,
+      "loss": 1.4686,
+      "step": 1550
+    },
+    {
+      "epoch": 3.565714285714286,
+      "grad_norm": 0.6395593881607056,
+      "learning_rate": 0.0002,
+      "loss": 1.4241,
+      "step": 1560
+    },
+    {
+      "epoch": 3.5885714285714285,
+      "grad_norm": 0.7588621377944946,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1570
+    },
+    {
+      "epoch": 3.611428571428571,
+      "grad_norm": 0.6206456422805786,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 1580
+    },
+    {
+      "epoch": 3.6342857142857143,
+      "grad_norm": 0.7591291666030884,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 1590
+    },
+    {
+      "epoch": 3.657142857142857,
+      "grad_norm": 0.6476313471794128,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 1600
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 0.6731392741203308,
+      "learning_rate": 0.0002,
+      "loss": 1.4598,
+      "step": 1610
+    },
+    {
+      "epoch": 3.702857142857143,
+      "grad_norm": 0.725190281867981,
+      "learning_rate": 0.0002,
+      "loss": 1.4225,
+      "step": 1620
+    },
+    {
+      "epoch": 3.725714285714286,
+      "grad_norm": 0.6720049977302551,
+      "learning_rate": 0.0002,
+      "loss": 1.4525,
+      "step": 1630
+    },
+    {
+      "epoch": 3.7485714285714287,
+      "grad_norm": 0.6301007270812988,
+      "learning_rate": 0.0002,
+      "loss": 1.429,
+      "step": 1640
+    },
+    {
+      "epoch": 3.7714285714285714,
+      "grad_norm": 0.715893566608429,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 1650
+    },
+    {
+      "epoch": 3.7942857142857145,
+      "grad_norm": 0.7539359927177429,
+      "learning_rate": 0.0002,
+      "loss": 1.3624,
+      "step": 1660
+    },
+    {
+      "epoch": 3.817142857142857,
+      "grad_norm": 0.6658543348312378,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 1670
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 0.7019526958465576,
+      "learning_rate": 0.0002,
+      "loss": 1.3934,
+      "step": 1680
+    },
+    {
+      "epoch": 3.862857142857143,
+      "grad_norm": 0.6517802476882935,
+      "learning_rate": 0.0002,
+      "loss": 1.4436,
+      "step": 1690
+    },
+    {
+      "epoch": 3.8857142857142857,
+      "grad_norm": 0.7617332935333252,
+      "learning_rate": 0.0002,
+      "loss": 1.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 3.9085714285714284,
+      "grad_norm": 0.6919480562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 1710
+    },
+    {
+      "epoch": 3.9314285714285715,
+      "grad_norm": 0.6987943053245544,
+      "learning_rate": 0.0002,
+      "loss": 1.4317,
+      "step": 1720
+    },
+    {
+      "epoch": 3.954285714285714,
+      "grad_norm": 0.7062228918075562,
+      "learning_rate": 0.0002,
+      "loss": 1.4704,
+      "step": 1730
+    },
+    {
+      "epoch": 3.977142857142857,
+      "grad_norm": 0.6769542098045349,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 1740
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6832144260406494,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 1750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9474865198135376,
+      "eval_runtime": 111.288,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 1750
+    },
+    {
+      "epoch": 4.022857142857143,
+      "grad_norm": 1.064110279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 1760
+    },
+    {
+      "epoch": 4.045714285714285,
+      "grad_norm": 0.8380683660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.2013,
+      "step": 1770
+    },
+    {
+      "epoch": 4.0685714285714285,
+      "grad_norm": 1.1863020658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 1780
+    },
+    {
+      "epoch": 4.091428571428572,
+      "grad_norm": 1.0128898620605469,
+      "learning_rate": 0.0002,
+      "loss": 1.2499,
+      "step": 1790
+    },
+    {
+      "epoch": 4.114285714285714,
+      "grad_norm": 0.9221312403678894,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 1800
+    },
+    {
+      "epoch": 4.137142857142857,
+      "grad_norm": 1.1298727989196777,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 1810
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 0.8854547739028931,
+      "learning_rate": 0.0002,
+      "loss": 1.1491,
+      "step": 1820
+    },
+    {
+      "epoch": 4.182857142857143,
+      "grad_norm": 0.8920808434486389,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1830
+    },
+    {
+      "epoch": 4.2057142857142855,
+      "grad_norm": 0.913244366645813,
+      "learning_rate": 0.0002,
+      "loss": 1.1969,
+      "step": 1840
+    },
+    {
+      "epoch": 4.228571428571429,
+      "grad_norm": 0.908831000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 1850
+    },
+    {
+      "epoch": 4.251428571428572,
+      "grad_norm": 1.0223685503005981,
+      "learning_rate": 0.0002,
+      "loss": 1.1653,
+      "step": 1860
+    },
+    {
+      "epoch": 4.274285714285714,
+      "grad_norm": 0.9771921634674072,
+      "learning_rate": 0.0002,
+      "loss": 1.2497,
+      "step": 1870
+    },
+    {
+      "epoch": 4.297142857142857,
+      "grad_norm": 0.9313384890556335,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 1880
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 1.0754257440567017,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 1890
+    },
+    {
+      "epoch": 4.3428571428571425,
+      "grad_norm": 0.8904672265052795,
+      "learning_rate": 0.0002,
+      "loss": 1.2286,
+      "step": 1900
+    },
+    {
+      "epoch": 4.365714285714286,
+      "grad_norm": 1.046527624130249,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 1910
+    },
+    {
+      "epoch": 4.388571428571429,
+      "grad_norm": 0.9576982855796814,
+      "learning_rate": 0.0002,
+      "loss": 1.2368,
+      "step": 1920
+    },
+    {
+      "epoch": 4.411428571428571,
+      "grad_norm": 0.9278356432914734,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 1930
+    },
+    {
+      "epoch": 4.434285714285714,
+      "grad_norm": 1.1763030290603638,
+      "learning_rate": 0.0002,
+      "loss": 1.2005,
+      "step": 1940
+    },
+    {
+      "epoch": 4.457142857142857,
+      "grad_norm": 0.9183000326156616,
+      "learning_rate": 0.0002,
+      "loss": 1.1541,
+      "step": 1950
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 1.050980806350708,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 1960
+    },
+    {
+      "epoch": 4.502857142857143,
+      "grad_norm": 0.9975392818450928,
+      "learning_rate": 0.0002,
+      "loss": 1.2133,
+      "step": 1970
+    },
+    {
+      "epoch": 4.525714285714286,
+      "grad_norm": 0.990544319152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2312,
+      "step": 1980
+    },
+    {
+      "epoch": 4.548571428571429,
+      "grad_norm": 1.004794955253601,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 1990
+    },
+    {
+      "epoch": 4.571428571428571,
+      "grad_norm": 0.9294857978820801,
+      "learning_rate": 0.0002,
+      "loss": 1.2085,
+      "step": 2000
+    },
+    {
+      "epoch": 4.594285714285714,
+      "grad_norm": 0.93436598777771,
+      "learning_rate": 0.0002,
+      "loss": 1.2874,
+      "step": 2010
+    },
+    {
+      "epoch": 4.617142857142857,
+      "grad_norm": 0.8704655766487122,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 2020
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 0.9077927470207214,
+      "learning_rate": 0.0002,
+      "loss": 1.204,
+      "step": 2030
+    },
+    {
+      "epoch": 4.662857142857143,
+      "grad_norm": 0.912987470626831,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 2040
+    },
+    {
+      "epoch": 4.685714285714286,
+      "grad_norm": 0.9740643501281738,
+      "learning_rate": 0.0002,
+      "loss": 1.2868,
+      "step": 2050
+    },
+    {
+      "epoch": 4.708571428571428,
+      "grad_norm": 1.133357048034668,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 2060
+    },
+    {
+      "epoch": 4.731428571428571,
+      "grad_norm": 0.8844527006149292,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 2070
+    },
+    {
+      "epoch": 4.7542857142857144,
+      "grad_norm": 1.0083311796188354,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 2080
+    },
+    {
+      "epoch": 4.777142857142858,
+      "grad_norm": 1.000447154045105,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 2090
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.9620300531387329,
+      "learning_rate": 0.0002,
+      "loss": 1.2313,
+      "step": 2100
+    },
+    {
+      "epoch": 4.822857142857143,
+      "grad_norm": 0.9843335151672363,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 2110
+    },
+    {
+      "epoch": 4.845714285714286,
+      "grad_norm": 0.9906681180000305,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 2120
+    },
+    {
+      "epoch": 4.868571428571428,
+      "grad_norm": 0.9544073939323425,
+      "learning_rate": 0.0002,
+      "loss": 1.2325,
+      "step": 2130
+    },
+    {
+      "epoch": 4.8914285714285715,
+      "grad_norm": 0.9392994046211243,
+      "learning_rate": 0.0002,
+      "loss": 1.284,
+      "step": 2140
+    },
+    {
+      "epoch": 4.914285714285715,
+      "grad_norm": 1.104519248008728,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2150
+    },
+    {
+      "epoch": 4.937142857142857,
+      "grad_norm": 0.9495956897735596,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 2160
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 0.9696287512779236,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 2170
+    },
+    {
+      "epoch": 4.982857142857143,
+      "grad_norm": 0.9933681488037109,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 2180
+    },
+    {
+      "epoch": 4.998857142857143,
+      "eval_loss": 2.099808692932129,
+      "eval_runtime": 111.2808,
+      "eval_samples_per_second": 4.556,
+      "eval_steps_per_second": 0.575,
+      "step": 2187
+    },
+    {
+      "epoch": 5.005714285714285,
+      "grad_norm": 0.9482853412628174,
+      "learning_rate": 0.0002,
+      "loss": 1.2183,
+      "step": 2190
+    },
+    {
+      "epoch": 5.0285714285714285,
+      "grad_norm": 1.6689555644989014,
+      "learning_rate": 0.0002,
+      "loss": 0.9898,
+      "step": 2200
+    },
+    {
+      "epoch": 5.051428571428572,
+      "grad_norm": 1.2019699811935425,
+      "learning_rate": 0.0002,
+      "loss": 0.9741,
+      "step": 2210
+    },
+    {
+      "epoch": 5.074285714285715,
+      "grad_norm": 1.535780429840088,
+      "learning_rate": 0.0002,
+      "loss": 0.9737,
+      "step": 2220
+    },
+    {
+      "epoch": 5.097142857142857,
+      "grad_norm": 1.2061309814453125,
+      "learning_rate": 0.0002,
+      "loss": 0.9494,
+      "step": 2230
+    },
+    {
+      "epoch": 5.12,
+      "grad_norm": 1.1898778676986694,
+      "learning_rate": 0.0002,
+      "loss": 0.9316,
+      "step": 2240
+    },
+    {
+      "epoch": 5.142857142857143,
+      "grad_norm": 1.158898949623108,
+      "learning_rate": 0.0002,
+      "loss": 1.002,
+      "step": 2250
+    },
+    {
+      "epoch": 5.1657142857142855,
+      "grad_norm": 1.370749592781067,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 2260
+    },
+    {
+      "epoch": 5.188571428571429,
+      "grad_norm": 1.314120888710022,
+      "learning_rate": 0.0002,
+      "loss": 0.9365,
+      "step": 2270
+    },
+    {
+      "epoch": 5.211428571428572,
+      "grad_norm": 1.2184966802597046,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 2280
+    },
+    {
+      "epoch": 5.234285714285714,
+      "grad_norm": 1.4833279848098755,
+      "learning_rate": 0.0002,
+      "loss": 0.9407,
+      "step": 2290
+    },
+    {
+      "epoch": 5.257142857142857,
+      "grad_norm": 1.3348219394683838,
+      "learning_rate": 0.0002,
+      "loss": 0.9635,
+      "step": 2300
+    },
+    {
+      "epoch": 5.28,
+      "grad_norm": 1.4166619777679443,
+      "learning_rate": 0.0002,
+      "loss": 1.0294,
+      "step": 2310
+    },
+    {
+      "epoch": 5.3028571428571425,
+      "grad_norm": 1.4539530277252197,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 2320
+    },
+    {
+      "epoch": 5.325714285714286,
+      "grad_norm": 1.4642518758773804,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 2330
+    },
+    {
+      "epoch": 5.348571428571429,
+      "grad_norm": 1.3938848972320557,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 2340
+    },
+    {
+      "epoch": 5.371428571428572,
+      "grad_norm": 1.1147894859313965,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 2350
+    },
+    {
+      "epoch": 5.394285714285714,
+      "grad_norm": 1.3465309143066406,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2360
+    },
+    {
+      "epoch": 5.417142857142857,
+      "grad_norm": 1.4788566827774048,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 2370
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 1.3808705806732178,
+      "learning_rate": 0.0002,
+      "loss": 0.9896,
+      "step": 2380
+    },
+    {
+      "epoch": 5.462857142857143,
+      "grad_norm": 1.2336329221725464,
+      "learning_rate": 0.0002,
+      "loss": 1.0279,
+      "step": 2390
+    },
+    {
+      "epoch": 5.485714285714286,
+      "grad_norm": 1.5445678234100342,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 2400
+    },
+    {
+      "epoch": 5.508571428571429,
+      "grad_norm": 1.107488989830017,
+      "learning_rate": 0.0002,
+      "loss": 0.9534,
+      "step": 2410
+    },
+    {
+      "epoch": 5.531428571428571,
+      "grad_norm": 1.39687979221344,
+      "learning_rate": 0.0002,
+      "loss": 1.0036,
+      "step": 2420
+    },
+    {
+      "epoch": 5.554285714285714,
+      "grad_norm": 1.3905695676803589,
+      "learning_rate": 0.0002,
+      "loss": 0.9959,
+      "step": 2430
+    },
+    {
+      "epoch": 5.577142857142857,
+      "grad_norm": 1.3772821426391602,
+      "learning_rate": 0.0002,
+      "loss": 0.9912,
+      "step": 2440
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 1.1661899089813232,
+      "learning_rate": 0.0002,
+      "loss": 0.9825,
+      "step": 2450
+    },
+    {
+      "epoch": 5.622857142857143,
+      "grad_norm": 1.2730463743209839,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 2460
+    },
+    {
+      "epoch": 5.645714285714286,
+      "grad_norm": 1.2251193523406982,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2470
+    },
+    {
+      "epoch": 5.668571428571429,
+      "grad_norm": 1.5454859733581543,
+      "learning_rate": 0.0002,
+      "loss": 1.079,
+      "step": 2480
+    },
+    {
+      "epoch": 5.691428571428571,
+      "grad_norm": 1.5405735969543457,
+      "learning_rate": 0.0002,
+      "loss": 1.0414,
+      "step": 2490
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.2555434703826904,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 2500
+    },
+    {
+      "epoch": 5.737142857142857,
+      "grad_norm": 1.3323487043380737,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 2510
+    },
+    {
+      "epoch": 5.76,
+      "grad_norm": 1.3106356859207153,
+      "learning_rate": 0.0002,
+      "loss": 1.051,
+      "step": 2520
+    },
+    {
+      "epoch": 5.782857142857143,
+      "grad_norm": 1.4832439422607422,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 2530
+    },
+    {
+      "epoch": 5.805714285714286,
+      "grad_norm": 1.1336562633514404,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 2540
+    },
+    {
+      "epoch": 5.828571428571428,
+      "grad_norm": 1.2434223890304565,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2550
+    },
+    {
+      "epoch": 5.851428571428571,
+      "grad_norm": 1.2825450897216797,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 2560
+    },
+    {
+      "epoch": 5.8742857142857146,
+      "grad_norm": 1.4373180866241455,
+      "learning_rate": 0.0002,
+      "loss": 1.0642,
+      "step": 2570
+    },
+    {
+      "epoch": 5.897142857142857,
+      "grad_norm": 1.435015320777893,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 2580
+    },
+    {
+      "epoch": 5.92,
+      "grad_norm": 1.4075653553009033,
+      "learning_rate": 0.0002,
+      "loss": 1.0272,
+      "step": 2590
+    },
+    {
+      "epoch": 5.942857142857143,
+      "grad_norm": 1.319630742073059,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 2600
+    },
+    {
+      "epoch": 5.965714285714286,
+      "grad_norm": 1.278330683708191,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 2610
+    },
+    {
+      "epoch": 5.988571428571428,
+      "grad_norm": 1.258158564567566,
+      "learning_rate": 0.0002,
+      "loss": 1.0766,
+      "step": 2620
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.3689301013946533,
+      "eval_runtime": 53.9067,
+      "eval_samples_per_second": 9.405,
+      "eval_steps_per_second": 1.187,
+      "step": 2625
+    },
+    {
+      "epoch": 6.011428571428572,
+      "grad_norm": 1.3128368854522705,
+      "learning_rate": 0.0002,
+      "loss": 0.9142,
+      "step": 2630
+    },
+    {
+      "epoch": 6.034285714285715,
+      "grad_norm": 1.4280474185943604,
+      "learning_rate": 0.0002,
+      "loss": 0.7716,
+      "step": 2640
+    },
+    {
+      "epoch": 6.057142857142857,
+      "grad_norm": 1.5061450004577637,
+      "learning_rate": 0.0002,
+      "loss": 0.7776,
+      "step": 2650
+    },
+    {
+      "epoch": 6.08,
+      "grad_norm": 1.6013342142105103,
+      "learning_rate": 0.0002,
+      "loss": 0.7707,
+      "step": 2660
+    },
+    {
+      "epoch": 6.102857142857143,
+      "grad_norm": 2.0107381343841553,
+      "learning_rate": 0.0002,
+      "loss": 0.7543,
+      "step": 2670
+    },
+    {
+      "epoch": 6.1257142857142854,
+      "grad_norm": 1.5010124444961548,
+      "learning_rate": 0.0002,
+      "loss": 0.747,
+      "step": 2680
+    },
+    {
+      "epoch": 6.148571428571429,
+      "grad_norm": 1.5222150087356567,
+      "learning_rate": 0.0002,
+      "loss": 0.7501,
+      "step": 2690
+    },
+    {
+      "epoch": 6.171428571428572,
+      "grad_norm": 1.5413103103637695,
+      "learning_rate": 0.0002,
+      "loss": 0.7712,
+      "step": 2700
+    },
+    {
+      "epoch": 6.194285714285714,
+      "grad_norm": 1.527140736579895,
+      "learning_rate": 0.0002,
+      "loss": 0.7,
+      "step": 2710
+    },
+    {
+      "epoch": 6.217142857142857,
+      "grad_norm": 1.9386590719223022,
+      "learning_rate": 0.0002,
+      "loss": 0.7539,
+      "step": 2720
+    },
+    {
+      "epoch": 6.24,
+      "grad_norm": 1.8115214109420776,
+      "learning_rate": 0.0002,
+      "loss": 0.7586,
+      "step": 2730
+    },
+    {
+      "epoch": 6.2628571428571425,
+      "grad_norm": 1.6221802234649658,
+      "learning_rate": 0.0002,
+      "loss": 0.7426,
+      "step": 2740
+    },
+    {
+      "epoch": 6.285714285714286,
+      "grad_norm": 1.6698768138885498,
+      "learning_rate": 0.0002,
+      "loss": 0.8002,
+      "step": 2750
+    },
+    {
+      "epoch": 6.308571428571429,
+      "grad_norm": 1.7960610389709473,
+      "learning_rate": 0.0002,
+      "loss": 0.7293,
+      "step": 2760
+    },
+    {
+      "epoch": 6.331428571428571,
+      "grad_norm": 1.32172429561615,
+      "learning_rate": 0.0002,
+      "loss": 0.7405,
+      "step": 2770
+    },
+    {
+      "epoch": 6.354285714285714,
+      "grad_norm": 1.7468090057373047,
+      "learning_rate": 0.0002,
+      "loss": 0.7198,
+      "step": 2780
+    },
+    {
+      "epoch": 6.377142857142857,
+      "grad_norm": 1.6777397394180298,
+      "learning_rate": 0.0002,
+      "loss": 0.76,
+      "step": 2790
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 1.6200671195983887,
+      "learning_rate": 0.0002,
+      "loss": 0.7879,
+      "step": 2800
+    },
+    {
+      "epoch": 6.422857142857143,
+      "grad_norm": 1.723505973815918,
+      "learning_rate": 0.0002,
+      "loss": 0.7807,
+      "step": 2810
+    },
+    {
+      "epoch": 6.445714285714286,
+      "grad_norm": 1.4945589303970337,
+      "learning_rate": 0.0002,
+      "loss": 0.8645,
+      "step": 2820
+    },
+    {
+      "epoch": 6.468571428571429,
+      "grad_norm": 1.666458010673523,
+      "learning_rate": 0.0002,
+      "loss": 0.809,
+      "step": 2830
+    },
+    {
+      "epoch": 6.491428571428571,
+      "grad_norm": 1.6586525440216064,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 2840
+    },
+    {
+      "epoch": 6.514285714285714,
+      "grad_norm": 1.7480043172836304,
+      "learning_rate": 0.0002,
+      "loss": 0.8062,
+      "step": 2850
+    },
+    {
+      "epoch": 6.537142857142857,
+      "grad_norm": 1.4605649709701538,
+      "learning_rate": 0.0002,
+      "loss": 0.7602,
+      "step": 2860
+    },
+    {
+      "epoch": 6.5600000000000005,
+      "grad_norm": 1.4841814041137695,
+      "learning_rate": 0.0002,
+      "loss": 0.8186,
+      "step": 2870
+    },
+    {
+      "epoch": 6.582857142857143,
+      "grad_norm": 1.4653114080429077,
+      "learning_rate": 0.0002,
+      "loss": 0.8156,
+      "step": 2880
+    },
+    {
+      "epoch": 6.605714285714286,
+      "grad_norm": 1.7266837358474731,
+      "learning_rate": 0.0002,
+      "loss": 0.8111,
+      "step": 2890
+    },
+    {
+      "epoch": 6.628571428571428,
+      "grad_norm": 1.4860098361968994,
+      "learning_rate": 0.0002,
+      "loss": 0.7644,
+      "step": 2900
+    },
+    {
+      "epoch": 6.651428571428571,
+      "grad_norm": 1.7177597284317017,
+      "learning_rate": 0.0002,
+      "loss": 0.7991,
+      "step": 2910
+    },
+    {
+      "epoch": 6.674285714285714,
+      "grad_norm": 1.6757104396820068,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 2920
+    },
+    {
+      "epoch": 6.6971428571428575,
+      "grad_norm": 1.5177433490753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 2930
+    },
+    {
+      "epoch": 6.72,
+      "grad_norm": 1.8073889017105103,
+      "learning_rate": 0.0002,
+      "loss": 0.7825,
+      "step": 2940
+    },
+    {
+      "epoch": 6.742857142857143,
+      "grad_norm": 1.72337007522583,
+      "learning_rate": 0.0002,
+      "loss": 0.8234,
+      "step": 2950
+    },
+    {
+      "epoch": 6.765714285714286,
+      "grad_norm": 1.6298240423202515,
+      "learning_rate": 0.0002,
+      "loss": 0.896,
+      "step": 2960
+    },
+    {
+      "epoch": 6.788571428571428,
+      "grad_norm": 1.6140344142913818,
+      "learning_rate": 0.0002,
+      "loss": 0.8252,
+      "step": 2970
+    },
+    {
+      "epoch": 6.811428571428571,
+      "grad_norm": 1.7180862426757812,
+      "learning_rate": 0.0002,
+      "loss": 0.8314,
+      "step": 2980
+    },
+    {
+      "epoch": 6.8342857142857145,
+      "grad_norm": 1.7589894533157349,
+      "learning_rate": 0.0002,
+      "loss": 0.7929,
+      "step": 2990
+    },
+    {
+      "epoch": 6.857142857142857,
+      "grad_norm": 1.780195713043213,
+      "learning_rate": 0.0002,
+      "loss": 0.828,
+      "step": 3000
+    },
+    {
+      "epoch": 6.88,
+      "grad_norm": 1.7182508707046509,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 3010
+    },
+    {
+      "epoch": 6.902857142857143,
+      "grad_norm": 1.6308406591415405,
+      "learning_rate": 0.0002,
+      "loss": 0.7964,
+      "step": 3020
+    },
+    {
+      "epoch": 6.925714285714285,
+      "grad_norm": 1.5080229043960571,
+      "learning_rate": 0.0002,
+      "loss": 0.8207,
+      "step": 3030
+    },
+    {
+      "epoch": 6.948571428571428,
+      "grad_norm": 1.623555064201355,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 3040
+    },
+    {
+      "epoch": 6.9714285714285715,
+      "grad_norm": 1.526054859161377,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 3050
+    },
+    {
+      "epoch": 6.994285714285715,
+      "grad_norm": 1.6671174764633179,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3060
+    },
+    {
+      "epoch": 6.998857142857143,
+      "eval_loss": 2.647613525390625,
+      "eval_runtime": 111.2255,
+      "eval_samples_per_second": 4.558,
+      "eval_steps_per_second": 0.575,
+      "step": 3062
+    },
+    {
+      "epoch": 7.017142857142857,
+      "grad_norm": 1.9154540300369263,
+      "learning_rate": 0.0002,
+      "loss": 0.6303,
+      "step": 3070
+    },
+    {
+      "epoch": 7.04,
+      "grad_norm": 2.1938717365264893,
+      "learning_rate": 0.0002,
+      "loss": 0.5254,
+      "step": 3080
+    },
+    {
+      "epoch": 7.062857142857143,
+      "grad_norm": 1.7861053943634033,
+      "learning_rate": 0.0002,
+      "loss": 0.5087,
+      "step": 3090
+    },
+    {
+      "epoch": 7.085714285714285,
+      "grad_norm": 2.096458911895752,
+      "learning_rate": 0.0002,
+      "loss": 0.5751,
+      "step": 3100
+    },
+    {
+      "epoch": 7.1085714285714285,
+      "grad_norm": 2.0057616233825684,
+      "learning_rate": 0.0002,
+      "loss": 0.5188,
+      "step": 3110
+    },
+    {
+      "epoch": 7.131428571428572,
+      "grad_norm": 1.7073354721069336,
+      "learning_rate": 0.0002,
+      "loss": 0.5544,
+      "step": 3120
+    },
+    {
+      "epoch": 7.154285714285714,
+      "grad_norm": 2.3477938175201416,
+      "learning_rate": 0.0002,
+      "loss": 0.6005,
+      "step": 3130
+    },
+    {
+      "epoch": 7.177142857142857,
+      "grad_norm": 2.0903899669647217,
+      "learning_rate": 0.0002,
+      "loss": 0.5303,
+      "step": 3140
+    },
+    {
+      "epoch": 7.2,
+      "grad_norm": 1.7363157272338867,
+      "learning_rate": 0.0002,
+      "loss": 0.5397,
+      "step": 3150
+    },
+    {
+      "epoch": 7.222857142857142,
+      "grad_norm": 2.0611023902893066,
+      "learning_rate": 0.0002,
+      "loss": 0.5491,
+      "step": 3160
+    },
+    {
+      "epoch": 7.2457142857142856,
+      "grad_norm": 2.404407501220703,
+      "learning_rate": 0.0002,
+      "loss": 0.5865,
+      "step": 3170
+    },
+    {
+      "epoch": 7.268571428571429,
+      "grad_norm": 2.1841039657592773,
+      "learning_rate": 0.0002,
+      "loss": 0.5743,
+      "step": 3180
+    },
+    {
+      "epoch": 7.291428571428572,
+      "grad_norm": 1.7582741975784302,
+      "learning_rate": 0.0002,
+      "loss": 0.6604,
+      "step": 3190
+    },
+    {
+      "epoch": 7.314285714285714,
+      "grad_norm": 1.8890602588653564,
+      "learning_rate": 0.0002,
+      "loss": 0.5697,
+      "step": 3200
+    },
+    {
+      "epoch": 7.337142857142857,
+      "grad_norm": 1.8433198928833008,
+      "learning_rate": 0.0002,
+      "loss": 0.5969,
+      "step": 3210
+    },
+    {
+      "epoch": 7.36,
+      "grad_norm": 1.652266263961792,
+      "learning_rate": 0.0002,
+      "loss": 0.6019,
+      "step": 3220
+    },
+    {
+      "epoch": 7.382857142857143,
+      "grad_norm": 1.914348840713501,
+      "learning_rate": 0.0002,
+      "loss": 0.5439,
+      "step": 3230
+    },
+    {
+      "epoch": 7.405714285714286,
+      "grad_norm": 1.7440582513809204,
+      "learning_rate": 0.0002,
+      "loss": 0.6198,
+      "step": 3240
+    },
+    {
+      "epoch": 7.428571428571429,
+      "grad_norm": 1.9745666980743408,
+      "learning_rate": 0.0002,
+      "loss": 0.6022,
+      "step": 3250
+    },
+    {
+      "epoch": 7.451428571428571,
+      "grad_norm": 1.6567715406417847,
+      "learning_rate": 0.0002,
+      "loss": 0.5512,
+      "step": 3260
+    },
+    {
+      "epoch": 7.474285714285714,
+      "grad_norm": 1.5239425897598267,
+      "learning_rate": 0.0002,
+      "loss": 0.5985,
+      "step": 3270
+    },
+    {
+      "epoch": 7.497142857142857,
+      "grad_norm": 2.0668740272521973,
+      "learning_rate": 0.0002,
+      "loss": 0.5944,
+      "step": 3280
+    },
+    {
+      "epoch": 7.52,
+      "grad_norm": 1.9551687240600586,
+      "learning_rate": 0.0002,
+      "loss": 0.5424,
+      "step": 3290
+    },
+    {
+      "epoch": 7.542857142857143,
+      "grad_norm": 2.276602268218994,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 3300
+    },
+    {
+      "epoch": 7.565714285714286,
+      "grad_norm": 1.9060227870941162,
+      "learning_rate": 0.0002,
+      "loss": 0.6183,
+      "step": 3310
+    },
+    {
+      "epoch": 7.588571428571429,
+      "grad_norm": 2.0276358127593994,
+      "learning_rate": 0.0002,
+      "loss": 0.6151,
+      "step": 3320
+    },
+    {
+      "epoch": 7.611428571428571,
+      "grad_norm": 2.037238121032715,
+      "learning_rate": 0.0002,
+      "loss": 0.5731,
+      "step": 3330
+    },
+    {
+      "epoch": 7.634285714285714,
+      "grad_norm": 2.0060055255889893,
+      "learning_rate": 0.0002,
+      "loss": 0.6522,
+      "step": 3340
+    },
+    {
+      "epoch": 7.6571428571428575,
+      "grad_norm": 1.8366512060165405,
+      "learning_rate": 0.0002,
+      "loss": 0.6275,
+      "step": 3350
+    },
+    {
+      "epoch": 7.68,
+      "grad_norm": 2.0789284706115723,
+      "learning_rate": 0.0002,
+      "loss": 0.6535,
+      "step": 3360
+    },
+    {
+      "epoch": 7.702857142857143,
+      "grad_norm": 2.137089490890503,
+      "learning_rate": 0.0002,
+      "loss": 0.6197,
+      "step": 3370
+    },
+    {
+      "epoch": 7.725714285714286,
+      "grad_norm": 1.829277753829956,
+      "learning_rate": 0.0002,
+      "loss": 0.6267,
+      "step": 3380
+    },
+    {
+      "epoch": 7.748571428571428,
+      "grad_norm": 1.9483778476715088,
+      "learning_rate": 0.0002,
+      "loss": 0.6567,
+      "step": 3390
+    },
+    {
+      "epoch": 7.771428571428571,
+      "grad_norm": 2.0347065925598145,
+      "learning_rate": 0.0002,
+      "loss": 0.6393,
+      "step": 3400
+    },
+    {
+      "epoch": 7.7942857142857145,
+      "grad_norm": 2.0142312049865723,
+      "learning_rate": 0.0002,
+      "loss": 0.662,
+      "step": 3410
+    },
+    {
+      "epoch": 7.817142857142857,
+      "grad_norm": 2.152569055557251,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 3420
+    },
+    {
+      "epoch": 7.84,
+      "grad_norm": 1.7300190925598145,
+      "learning_rate": 0.0002,
+      "loss": 0.6805,
+      "step": 3430
+    },
+    {
+      "epoch": 7.862857142857143,
+      "grad_norm": 2.3944954872131348,
+      "learning_rate": 0.0002,
+      "loss": 0.6538,
+      "step": 3440
+    },
+    {
+      "epoch": 7.885714285714286,
+      "grad_norm": 2.1004269123077393,
+      "learning_rate": 0.0002,
+      "loss": 0.6187,
+      "step": 3450
+    },
+    {
+      "epoch": 7.908571428571428,
+      "grad_norm": 2.05513072013855,
+      "learning_rate": 0.0002,
+      "loss": 0.6626,
+      "step": 3460
+    },
+    {
+      "epoch": 7.9314285714285715,
+      "grad_norm": 1.9822633266448975,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 3470
+    },
+    {
+      "epoch": 7.954285714285715,
+      "grad_norm": 1.9649063348770142,
+      "learning_rate": 0.0002,
+      "loss": 0.6663,
+      "step": 3480
+    },
+    {
+      "epoch": 7.977142857142857,
+      "grad_norm": 1.7002657651901245,
+      "learning_rate": 0.0002,
+      "loss": 0.6192,
+      "step": 3490
+    },
+    {
+      "epoch": 7.990857142857143,
+      "eval_loss": 3.0625548362731934,
+      "eval_runtime": 53.8832,
+      "eval_samples_per_second": 9.409,
+      "eval_steps_per_second": 1.188,
+      "step": 3496
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 1.7956765471604736e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-3496/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fdedd2597ed628b316d57a658eb47e79e78b3be0
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35609dd580a3315061cc133ca3b6c33456f1569461f910179abb235a198c055c
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3826f1b6568ff0b417422be811b643e8db3fa8b9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b40ef2a31487bcb182b56d8976aa361fa65f75e01cf80a4077fe981e4b31ae7
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fd4d4e551aac7509b0d6c639582b1f4009777fc4
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ae52459ca240d2c9a269e617193c16cbe0b127c34a4272b142d33e442e92ea3
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0517455a6d3ba3b19b750a04281120ba36256c6a
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5bba0f0dd0346fd33c336ada199442da884b97bac60b1d0fab07049a506ae05
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3aa815973df0fc3a362a201a3f0fbe67053f9e78
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/trainer_state.json
@@ -0,0 +1,342 @@
+{
+  "best_metric": 1.8310153484344482,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437",
+  "epoch": 0.9988571428571429,
+  "eval_steps": 10,
+  "global_step": 437,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "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": 2.247163871232e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8b971d2d3728c073474494167a8733e971e412cf
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a973c78893d5c5f843df8157e4f5f5d58bd4afd185281397acca339964604cf
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c85437aafa0232354eb71a3750bfc543cac9f55e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3b3bf45f015bfc673da78e4ffff095a436715fa4f9e50945493f644b659ee6d
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6eae44d3e8693f5b1418d6e94ab297937fbf90ad
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9ed95d68c1bbf14501d3029a46f035f89ff87248d2476aef8f988288c811c7
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e4e46ab564a8a045362ac55dc1e7edd5847d63a0
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af1fa39639e63e8234687469d9a055a8928be9f72585a801dd0078003c362e0d
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..f7eb452ecd602d25a35ee2fe60aa84ee17cbede0
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/trainer_state.json
@@ -0,0 +1,658 @@
+{
+  "best_metric": 1.8279441595077515,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 875,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022857142857142857,
+      "grad_norm": 0.6273946762084961,
+      "learning_rate": 0.0002,
+      "loss": 3.066,
+      "step": 10
+    },
+    {
+      "epoch": 0.045714285714285714,
+      "grad_norm": 0.5300710201263428,
+      "learning_rate": 0.0002,
+      "loss": 2.5855,
+      "step": 20
+    },
+    {
+      "epoch": 0.06857142857142857,
+      "grad_norm": 0.6162196397781372,
+      "learning_rate": 0.0002,
+      "loss": 2.1887,
+      "step": 30
+    },
+    {
+      "epoch": 0.09142857142857143,
+      "grad_norm": 0.5143047571182251,
+      "learning_rate": 0.0002,
+      "loss": 2.1164,
+      "step": 40
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.4000673294067383,
+      "learning_rate": 0.0002,
+      "loss": 1.943,
+      "step": 50
+    },
+    {
+      "epoch": 0.13714285714285715,
+      "grad_norm": 0.444892555475235,
+      "learning_rate": 0.0002,
+      "loss": 1.9531,
+      "step": 60
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.4871707558631897,
+      "learning_rate": 0.0002,
+      "loss": 1.9435,
+      "step": 70
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": 0.451060026884079,
+      "learning_rate": 0.0002,
+      "loss": 1.9072,
+      "step": 80
+    },
+    {
+      "epoch": 0.2057142857142857,
+      "grad_norm": 0.3939569592475891,
+      "learning_rate": 0.0002,
+      "loss": 1.9312,
+      "step": 90
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.5033721923828125,
+      "learning_rate": 0.0002,
+      "loss": 1.8982,
+      "step": 100
+    },
+    {
+      "epoch": 0.25142857142857145,
+      "grad_norm": 0.3636534512042999,
+      "learning_rate": 0.0002,
+      "loss": 1.9148,
+      "step": 110
+    },
+    {
+      "epoch": 0.2742857142857143,
+      "grad_norm": 0.4391206204891205,
+      "learning_rate": 0.0002,
+      "loss": 1.8462,
+      "step": 120
+    },
+    {
+      "epoch": 0.29714285714285715,
+      "grad_norm": 0.5243169665336609,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 130
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4055655598640442,
+      "learning_rate": 0.0002,
+      "loss": 1.8875,
+      "step": 140
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.39735132455825806,
+      "learning_rate": 0.0002,
+      "loss": 1.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.3657142857142857,
+      "grad_norm": 0.4696349501609802,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 160
+    },
+    {
+      "epoch": 0.38857142857142857,
+      "grad_norm": 0.3987901508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.7566,
+      "step": 170
+    },
+    {
+      "epoch": 0.4114285714285714,
+      "grad_norm": 0.32404327392578125,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 180
+    },
+    {
+      "epoch": 0.4342857142857143,
+      "grad_norm": 0.3692261576652527,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 190
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.37267744541168213,
+      "learning_rate": 0.0002,
+      "loss": 1.823,
+      "step": 200
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3559934198856354,
+      "learning_rate": 0.0002,
+      "loss": 1.8564,
+      "step": 210
+    },
+    {
+      "epoch": 0.5028571428571429,
+      "grad_norm": 0.3374815285205841,
+      "learning_rate": 0.0002,
+      "loss": 1.8651,
+      "step": 220
+    },
+    {
+      "epoch": 0.5257142857142857,
+      "grad_norm": 0.34598177671432495,
+      "learning_rate": 0.0002,
+      "loss": 1.8683,
+      "step": 230
+    },
+    {
+      "epoch": 0.5485714285714286,
+      "grad_norm": 0.35629919171333313,
+      "learning_rate": 0.0002,
+      "loss": 1.8554,
+      "step": 240
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.3586862087249756,
+      "learning_rate": 0.0002,
+      "loss": 1.8751,
+      "step": 250
+    },
+    {
+      "epoch": 0.5942857142857143,
+      "grad_norm": 0.3198927342891693,
+      "learning_rate": 0.0002,
+      "loss": 1.7942,
+      "step": 260
+    },
+    {
+      "epoch": 0.6171428571428571,
+      "grad_norm": 0.37690025568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.81,
+      "step": 270
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.2855667471885681,
+      "learning_rate": 0.0002,
+      "loss": 1.8258,
+      "step": 280
+    },
+    {
+      "epoch": 0.6628571428571428,
+      "grad_norm": 0.3242695927619934,
+      "learning_rate": 0.0002,
+      "loss": 1.8288,
+      "step": 290
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.2960120141506195,
+      "learning_rate": 0.0002,
+      "loss": 1.7738,
+      "step": 300
+    },
+    {
+      "epoch": 0.7085714285714285,
+      "grad_norm": 0.3596384823322296,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 310
+    },
+    {
+      "epoch": 0.7314285714285714,
+      "grad_norm": 0.3001834750175476,
+      "learning_rate": 0.0002,
+      "loss": 1.8132,
+      "step": 320
+    },
+    {
+      "epoch": 0.7542857142857143,
+      "grad_norm": 0.31361159682273865,
+      "learning_rate": 0.0002,
+      "loss": 1.7751,
+      "step": 330
+    },
+    {
+      "epoch": 0.7771428571428571,
+      "grad_norm": 0.34093308448791504,
+      "learning_rate": 0.0002,
+      "loss": 1.8021,
+      "step": 340
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.3383876085281372,
+      "learning_rate": 0.0002,
+      "loss": 1.7674,
+      "step": 350
+    },
+    {
+      "epoch": 0.8228571428571428,
+      "grad_norm": 0.35100996494293213,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 360
+    },
+    {
+      "epoch": 0.8457142857142858,
+      "grad_norm": 0.344976007938385,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 370
+    },
+    {
+      "epoch": 0.8685714285714285,
+      "grad_norm": 0.3119729459285736,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 380
+    },
+    {
+      "epoch": 0.8914285714285715,
+      "grad_norm": 0.349221795797348,
+      "learning_rate": 0.0002,
+      "loss": 1.8043,
+      "step": 390
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.3124293386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 400
+    },
+    {
+      "epoch": 0.9371428571428572,
+      "grad_norm": 0.35504350066185,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 410
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.310310959815979,
+      "learning_rate": 0.0002,
+      "loss": 1.8115,
+      "step": 420
+    },
+    {
+      "epoch": 0.9828571428571429,
+      "grad_norm": 0.30432847142219543,
+      "learning_rate": 0.0002,
+      "loss": 1.7666,
+      "step": 430
+    },
+    {
+      "epoch": 0.9988571428571429,
+      "eval_loss": 1.8310153484344482,
+      "eval_runtime": 111.7814,
+      "eval_samples_per_second": 4.536,
+      "eval_steps_per_second": 0.573,
+      "step": 437
+    },
+    {
+      "epoch": 1.0057142857142858,
+      "grad_norm": 0.3121616840362549,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 440
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3365118205547333,
+      "learning_rate": 0.0002,
+      "loss": 1.7404,
+      "step": 450
+    },
+    {
+      "epoch": 1.0514285714285714,
+      "grad_norm": 0.3626686930656433,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 460
+    },
+    {
+      "epoch": 1.0742857142857143,
+      "grad_norm": 0.30539533495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.721,
+      "step": 470
+    },
+    {
+      "epoch": 1.0971428571428572,
+      "grad_norm": 0.3159816861152649,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 480
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.3695855736732483,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 490
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3609161674976349,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 500
+    },
+    {
+      "epoch": 1.1657142857142857,
+      "grad_norm": 0.3683869242668152,
+      "learning_rate": 0.0002,
+      "loss": 1.8723,
+      "step": 510
+    },
+    {
+      "epoch": 1.1885714285714286,
+      "grad_norm": 0.3862539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.7102,
+      "step": 520
+    },
+    {
+      "epoch": 1.2114285714285715,
+      "grad_norm": 0.4244740307331085,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 530
+    },
+    {
+      "epoch": 1.2342857142857142,
+      "grad_norm": 0.373703271150589,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 540
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.35715773701667786,
+      "learning_rate": 0.0002,
+      "loss": 1.795,
+      "step": 550
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.3555964231491089,
+      "learning_rate": 0.0002,
+      "loss": 1.7578,
+      "step": 560
+    },
+    {
+      "epoch": 1.302857142857143,
+      "grad_norm": 0.35080263018608093,
+      "learning_rate": 0.0002,
+      "loss": 1.7228,
+      "step": 570
+    },
+    {
+      "epoch": 1.3257142857142856,
+      "grad_norm": 0.3589482307434082,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 580
+    },
+    {
+      "epoch": 1.3485714285714285,
+      "grad_norm": 0.3711223900318146,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 590
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.313614159822464,
+      "learning_rate": 0.0002,
+      "loss": 1.7417,
+      "step": 600
+    },
+    {
+      "epoch": 1.3942857142857144,
+      "grad_norm": 0.3842357397079468,
+      "learning_rate": 0.0002,
+      "loss": 1.7191,
+      "step": 610
+    },
+    {
+      "epoch": 1.4171428571428573,
+      "grad_norm": 0.36126819252967834,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 620
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.35922661423683167,
+      "learning_rate": 0.0002,
+      "loss": 1.7073,
+      "step": 630
+    },
+    {
+      "epoch": 1.4628571428571429,
+      "grad_norm": 0.3922875225543976,
+      "learning_rate": 0.0002,
+      "loss": 1.6708,
+      "step": 640
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.365546852350235,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 650
+    },
+    {
+      "epoch": 1.5085714285714285,
+      "grad_norm": 0.36107590794563293,
+      "learning_rate": 0.0002,
+      "loss": 1.674,
+      "step": 660
+    },
+    {
+      "epoch": 1.5314285714285716,
+      "grad_norm": 0.3307042121887207,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 670
+    },
+    {
+      "epoch": 1.5542857142857143,
+      "grad_norm": 0.3492133915424347,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 680
+    },
+    {
+      "epoch": 1.5771428571428572,
+      "grad_norm": 0.38608574867248535,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 690
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.3489173650741577,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 700
+    },
+    {
+      "epoch": 1.6228571428571428,
+      "grad_norm": 0.36614152789115906,
+      "learning_rate": 0.0002,
+      "loss": 1.7306,
+      "step": 710
+    },
+    {
+      "epoch": 1.6457142857142857,
+      "grad_norm": 0.34340205788612366,
+      "learning_rate": 0.0002,
+      "loss": 1.7281,
+      "step": 720
+    },
+    {
+      "epoch": 1.6685714285714286,
+      "grad_norm": 0.34590771794319153,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 730
+    },
+    {
+      "epoch": 1.6914285714285713,
+      "grad_norm": 0.3759954273700714,
+      "learning_rate": 0.0002,
+      "loss": 1.7108,
+      "step": 740
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.3753475546836853,
+      "learning_rate": 0.0002,
+      "loss": 1.6903,
+      "step": 750
+    },
+    {
+      "epoch": 1.737142857142857,
+      "grad_norm": 0.38416001200675964,
+      "learning_rate": 0.0002,
+      "loss": 1.7054,
+      "step": 760
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 0.36223554611206055,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 770
+    },
+    {
+      "epoch": 1.782857142857143,
+      "grad_norm": 0.329556941986084,
+      "learning_rate": 0.0002,
+      "loss": 1.6878,
+      "step": 780
+    },
+    {
+      "epoch": 1.8057142857142856,
+      "grad_norm": 0.34008052945137024,
+      "learning_rate": 0.0002,
+      "loss": 1.803,
+      "step": 790
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.40297919511795044,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 800
+    },
+    {
+      "epoch": 1.8514285714285714,
+      "grad_norm": 0.35378390550613403,
+      "learning_rate": 0.0002,
+      "loss": 1.7518,
+      "step": 810
+    },
+    {
+      "epoch": 1.8742857142857143,
+      "grad_norm": 0.3625478148460388,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 820
+    },
+    {
+      "epoch": 1.8971428571428572,
+      "grad_norm": 0.36153221130371094,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 830
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 0.3612948954105377,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 840
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.399213045835495,
+      "learning_rate": 0.0002,
+      "loss": 1.7242,
+      "step": 850
+    },
+    {
+      "epoch": 1.9657142857142857,
+      "grad_norm": 0.40026402473449707,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 860
+    },
+    {
+      "epoch": 1.9885714285714284,
+      "grad_norm": 0.38114118576049805,
+      "learning_rate": 0.0002,
+      "loss": 1.76,
+      "step": 870
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8279441595077515,
+      "eval_runtime": 111.1652,
+      "eval_samples_per_second": 4.561,
+      "eval_steps_per_second": 0.576,
+      "step": 875
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3496,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.494327742464e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6e8dec0ead2c8272aa4a091c380e4689f369a93e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b24d137580f566a690ee384c6a9afca9277476edf438c9aac06f28988d9d164
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_log.jsonl b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..275174b2c5baca4834779650122bfe11bcce1186
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/training_log.jsonl
@@ -0,0 +1,13 @@
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 2864.7610309123993, "total_accumulated_duration": 2864.7610309123993, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0595, "grad_norm": 0.8717339634895325, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5797, "grad_norm": 0.5135980248451233, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1883, "grad_norm": 0.5807254910469055, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1197, "grad_norm": 0.5022327303886414, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.9424, "grad_norm": 0.39211124181747437, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9487, "grad_norm": 0.4291403591632843, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9404, "grad_norm": 0.43344980478286743, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9033, "grad_norm": 0.45926913619041443, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9333, "grad_norm": 0.46012991666793823, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8955, "grad_norm": 0.5238164067268372, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9122, "grad_norm": 0.39393168687820435, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8442, "grad_norm": 0.3722357749938965, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7979, "grad_norm": 0.4925648272037506, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8865, "grad_norm": 0.3977905213832855, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.834, "grad_norm": 0.34175440669059753, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.8256, "grad_norm": 0.4174216389656067, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7573, "grad_norm": 0.3641279339790344, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.31452980637550354, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8204, "grad_norm": 0.30585241317749023, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.8247, "grad_norm": 0.35243943333625793, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8546, "grad_norm": 0.37146371603012085, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.865, "grad_norm": 0.3326093852519989, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8701, "grad_norm": 0.3610652983188629, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8517, "grad_norm": 0.37059760093688965, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8752, "grad_norm": 0.348164826631546, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7926, "grad_norm": 0.32422614097595215, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.8074, "grad_norm": 0.3642662465572357, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8236, "grad_norm": 0.2940135896205902, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8289, "grad_norm": 0.31184637546539307, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7761, "grad_norm": 0.31996026635169983, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8445, "grad_norm": 0.3466435372829437, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8114, "grad_norm": 0.29803967475891113, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7734, "grad_norm": 0.30836978554725647, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8042, "grad_norm": 0.3359459936618805, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7711, "grad_norm": 0.33156177401542664, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.7289, "grad_norm": 0.34951063990592957, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8496, "grad_norm": 0.3386661112308502, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8789, "grad_norm": 0.3092683255672455, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8021, "grad_norm": 0.3764317035675049, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7513, "grad_norm": 0.32476580142974854, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7613, "grad_norm": 0.33311036229133606, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8133, "grad_norm": 0.29290494322776794, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7675, "grad_norm": 0.30328553915023804, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 680.2093601226807, "total_accumulated_duration": 680.2093601226807, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0737, "grad_norm": 0.676007866859436, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5931, "grad_norm": 0.5250661969184875, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.2149, "grad_norm": 0.5478432774543762, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1497, "grad_norm": 0.4776158034801483, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.9594, "grad_norm": 0.42377957701683044, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9602, "grad_norm": 0.4962829649448395, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9458, "grad_norm": 0.4248361587524414, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9088, "grad_norm": 0.4115945100784302, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9341, "grad_norm": 0.37363702058792114, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8998, "grad_norm": 0.42065081000328064, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9176, "grad_norm": 0.37641531229019165, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8522, "grad_norm": 0.41112732887268066, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7993, "grad_norm": 0.3546443283557892, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8909, "grad_norm": 0.38930970430374146, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8342, "grad_norm": 0.35351401567459106, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.8289, "grad_norm": 0.4080873727798462, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7552, "grad_norm": 0.3885005712509155, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8296, "grad_norm": 0.3329996168613434, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8208, "grad_norm": 0.3085007965564728, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.8251, "grad_norm": 0.34996384382247925, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8534, "grad_norm": 0.35104790329933167, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8658, "grad_norm": 0.40699303150177, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.873, "grad_norm": 0.3361871838569641, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8498, "grad_norm": 0.3529999554157257, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8757, "grad_norm": 0.3695283532142639, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7928, "grad_norm": 0.32378143072128296, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.8093, "grad_norm": 0.38105419278144836, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8238, "grad_norm": 0.29338300228118896, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8324, "grad_norm": 0.3283067047595978, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7762, "grad_norm": 0.31334781646728516, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8455, "grad_norm": 0.32596927881240845, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8107, "grad_norm": 0.3007946312427521, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7752, "grad_norm": 0.31740236282348633, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8079, "grad_norm": 0.357985258102417, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7742, "grad_norm": 0.36722445487976074, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.7309, "grad_norm": 0.3409443497657776, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8493, "grad_norm": 0.35894039273262024, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8798, "grad_norm": 0.3310607969760895, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8057, "grad_norm": 0.3114466071128845, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7537, "grad_norm": 0.30708882212638855, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7645, "grad_norm": 0.33819758892059326, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8153, "grad_norm": 0.30491453409194946, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7678, "grad_norm": 0.3099863529205322, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 1692.0529749393463, "total_accumulated_duration": 1692.0529749393463, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0635, "grad_norm": 0.751778781414032, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5861, "grad_norm": 0.5179391503334045, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.204, "grad_norm": 0.588005006313324, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1343, "grad_norm": 0.4717636704444885, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.9485, "grad_norm": 0.42742547392845154, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9543, "grad_norm": 0.46707019209861755, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9416, "grad_norm": 0.4594128429889679, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9003, "grad_norm": 0.4940229654312134, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9331, "grad_norm": 0.3711043894290924, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8954, "grad_norm": 0.48097744584083557, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9152, "grad_norm": 0.36268872022628784, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8476, "grad_norm": 0.40977758169174194, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.8017, "grad_norm": 0.44437164068222046, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8891, "grad_norm": 0.4007284343242645, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8344, "grad_norm": 0.35068053007125854, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.8242, "grad_norm": 0.3963766098022461, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7578, "grad_norm": 0.34578168392181396, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.3284793794155121, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8207, "grad_norm": 0.33896616101264954, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.8213, "grad_norm": 0.3573452830314636, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8525, "grad_norm": 0.3636205792427063, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8663, "grad_norm": 0.3481282889842987, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8691, "grad_norm": 0.3532394766807556, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8508, "grad_norm": 0.34349769353866577, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8735, "grad_norm": 0.34332799911499023, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7917, "grad_norm": 0.3243716359138489, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.8104, "grad_norm": 0.36064547300338745, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8232, "grad_norm": 0.28934213519096375, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.833, "grad_norm": 0.30864956974983215, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.775, "grad_norm": 0.2911549210548401, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8431, "grad_norm": 0.3338742256164551, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8111, "grad_norm": 0.30065304040908813, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7735, "grad_norm": 0.31962037086486816, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8013, "grad_norm": 0.3339603543281555, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7657, "grad_norm": 0.36495035886764526, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.7268, "grad_norm": 1.9275314807891846, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8487, "grad_norm": 0.3443487882614136, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8788, "grad_norm": 0.32266315817832947, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8039, "grad_norm": 0.366872102022171, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7528, "grad_norm": 0.31905823945999146, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7587, "grad_norm": 0.3308110535144806, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8102, "grad_norm": 0.2978809177875519, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.768, "grad_norm": 0.3174574673175812, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 2183.624102115631, "total_accumulated_duration": 2183.624102115631, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0688, "grad_norm": 0.7712277770042419, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5869, "grad_norm": 0.526583194732666, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1872, "grad_norm": 0.6245124936103821, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1127, "grad_norm": 0.5284358859062195, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.9428, "grad_norm": 0.40955156087875366, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9539, "grad_norm": 0.43062952160835266, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9474, "grad_norm": 0.43123340606689453, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9062, "grad_norm": 0.44926217198371887, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9365, "grad_norm": 0.41470426321029663, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.895, "grad_norm": 0.4223814010620117, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9131, "grad_norm": 0.3610304296016693, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8466, "grad_norm": 0.4295799434185028, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.804, "grad_norm": 0.45776671171188354, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4016251564025879, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8378, "grad_norm": 0.48137062788009644, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.8227, "grad_norm": 0.49422022700309753, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7536, "grad_norm": 0.34034672379493713, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8266, "grad_norm": 0.3166326582431793, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8211, "grad_norm": 0.3084140717983246, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.821, "grad_norm": 0.3594462275505066, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.854, "grad_norm": 0.3633425831794739, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8614, "grad_norm": 0.356911838054657, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8722, "grad_norm": 0.47076764702796936, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8526, "grad_norm": 0.3551470935344696, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8764, "grad_norm": 0.3632679581642151, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7932, "grad_norm": 0.32329922914505005, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.8088, "grad_norm": 0.3659713566303253, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8236, "grad_norm": 0.30823490023612976, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8291, "grad_norm": 0.31695812940597534, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7757, "grad_norm": 0.30165690183639526, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8446, "grad_norm": 0.3283828794956207, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8093, "grad_norm": 0.3035045266151428, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7727, "grad_norm": 0.3053770661354065, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8035, "grad_norm": 0.3487062156200409, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7687, "grad_norm": 0.34441572427749634, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.7269, "grad_norm": 0.35523343086242676, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8515, "grad_norm": 0.3516801595687866, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8815, "grad_norm": 0.31675922870635986, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8073, "grad_norm": 0.3547166883945465, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7517, "grad_norm": 0.3162825405597687, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.765, "grad_norm": 0.3345049023628235, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8113, "grad_norm": 0.30412253737449646, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7697, "grad_norm": 0.3209483027458191, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 1455.278377532959, "total_accumulated_duration": 1455.278377532959, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 2.0, "step": 875, "epoch_duration": 1423.373628616333, "total_accumulated_duration": 2878.652006149292, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-437", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}]}
+{"epoch": 2.998857142857143, "step": 1312, "epoch_duration": 1505.2112596035004, "total_accumulated_duration": 4383.863265752792, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}]}
+{"epoch": 4.0, "step": 1750, "epoch_duration": 1378.1591472625732, "total_accumulated_duration": 5762.022413015366, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}, {"eval_loss": 1.8684407472610474, "eval_runtime": 111.2835, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 2.998857142857143, "step": 1312}, {"loss": 1.4486, "grad_norm": 0.5184893012046814, "learning_rate": 0.0002, "epoch": 3.0171428571428573, "step": 1320}, {"loss": 1.4082, "grad_norm": 0.5665355920791626, "learning_rate": 0.0002, "epoch": 3.04, "step": 1330}, {"loss": 1.3741, "grad_norm": 0.6601403951644897, "learning_rate": 0.0002, "epoch": 3.0628571428571427, "step": 1340}, {"loss": 1.433, "grad_norm": 0.6921621561050415, "learning_rate": 0.0002, "epoch": 3.085714285714286, "step": 1350}, {"loss": 1.4562, "grad_norm": 0.6406348943710327, "learning_rate": 0.0002, "epoch": 3.1085714285714285, "step": 1360}, {"loss": 1.3563, "grad_norm": 0.5814554691314697, "learning_rate": 0.0002, "epoch": 3.1314285714285712, "step": 1370}, {"loss": 1.4096, "grad_norm": 0.683325469493866, "learning_rate": 0.0002, "epoch": 3.1542857142857144, "step": 1380}, {"loss": 1.4106, "grad_norm": 0.6686155200004578, "learning_rate": 0.0002, "epoch": 3.177142857142857, "step": 1390}, {"loss": 1.4394, "grad_norm": 0.8159713745117188, "learning_rate": 0.0002, "epoch": 3.2, "step": 1400}, {"loss": 1.4279, "grad_norm": 0.646216094493866, "learning_rate": 0.0002, "epoch": 3.222857142857143, "step": 1410}, {"loss": 1.4232, "grad_norm": 0.7323529720306396, "learning_rate": 0.0002, "epoch": 3.2457142857142856, "step": 1420}, {"loss": 1.3891, "grad_norm": 0.689349353313446, "learning_rate": 0.0002, "epoch": 3.2685714285714287, "step": 1430}, {"loss": 1.4578, "grad_norm": 0.727894127368927, "learning_rate": 0.0002, "epoch": 3.2914285714285714, "step": 1440}, {"loss": 1.4, "grad_norm": 0.6921590566635132, "learning_rate": 0.0002, "epoch": 3.314285714285714, "step": 1450}, {"loss": 1.4272, "grad_norm": 0.6176243424415588, "learning_rate": 0.0002, "epoch": 3.337142857142857, "step": 1460}, {"loss": 1.4323, "grad_norm": 0.9006354212760925, "learning_rate": 0.0002, "epoch": 3.36, "step": 1470}, {"loss": 1.4353, "grad_norm": 0.8145929574966431, "learning_rate": 0.0002, "epoch": 3.382857142857143, "step": 1480}, {"loss": 1.3859, "grad_norm": 0.6640016436576843, "learning_rate": 0.0002, "epoch": 3.4057142857142857, "step": 1490}, {"loss": 1.387, "grad_norm": 0.7266780138015747, "learning_rate": 0.0002, "epoch": 3.4285714285714284, "step": 1500}, {"loss": 1.4108, "grad_norm": 0.9351356029510498, "learning_rate": 0.0002, "epoch": 3.4514285714285715, "step": 1510}, {"loss": 1.4656, "grad_norm": 0.675645649433136, "learning_rate": 0.0002, "epoch": 3.474285714285714, "step": 1520}, {"loss": 1.384, "grad_norm": 0.761472225189209, "learning_rate": 0.0002, "epoch": 3.4971428571428573, "step": 1530}, {"loss": 1.4968, "grad_norm": 0.6653069257736206, "learning_rate": 0.0002, "epoch": 3.52, "step": 1540}, {"loss": 1.4686, "grad_norm": 0.667412519454956, "learning_rate": 0.0002, "epoch": 3.5428571428571427, "step": 1550}, {"loss": 1.4241, "grad_norm": 0.6395593881607056, "learning_rate": 0.0002, "epoch": 3.565714285714286, "step": 1560}, {"loss": 1.4825, "grad_norm": 0.7588621377944946, "learning_rate": 0.0002, "epoch": 3.5885714285714285, "step": 1570}, {"loss": 1.4459, "grad_norm": 0.6206456422805786, "learning_rate": 0.0002, "epoch": 3.611428571428571, "step": 1580}, {"loss": 1.436, "grad_norm": 0.7591291666030884, "learning_rate": 0.0002, "epoch": 3.6342857142857143, "step": 1590}, {"loss": 1.458, "grad_norm": 0.6476313471794128, "learning_rate": 0.0002, "epoch": 3.657142857142857, "step": 1600}, {"loss": 1.4598, "grad_norm": 0.6731392741203308, "learning_rate": 0.0002, "epoch": 3.68, "step": 1610}, {"loss": 1.4225, "grad_norm": 0.725190281867981, "learning_rate": 0.0002, "epoch": 3.702857142857143, "step": 1620}, {"loss": 1.4525, "grad_norm": 0.6720049977302551, "learning_rate": 0.0002, "epoch": 3.725714285714286, "step": 1630}, {"loss": 1.429, "grad_norm": 0.6301007270812988, "learning_rate": 0.0002, "epoch": 3.7485714285714287, "step": 1640}, {"loss": 1.4166, "grad_norm": 0.715893566608429, "learning_rate": 0.0002, "epoch": 3.7714285714285714, "step": 1650}, {"loss": 1.3624, "grad_norm": 0.7539359927177429, "learning_rate": 0.0002, "epoch": 3.7942857142857145, "step": 1660}, {"loss": 1.4516, "grad_norm": 0.6658543348312378, "learning_rate": 0.0002, "epoch": 3.817142857142857, "step": 1670}, {"loss": 1.3934, "grad_norm": 0.7019526958465576, "learning_rate": 0.0002, "epoch": 3.84, "step": 1680}, {"loss": 1.4436, "grad_norm": 0.6517802476882935, "learning_rate": 0.0002, "epoch": 3.862857142857143, "step": 1690}, {"loss": 1.4968, "grad_norm": 0.7617332935333252, "learning_rate": 0.0002, "epoch": 3.8857142857142857, "step": 1700}, {"loss": 1.5145, "grad_norm": 0.6919480562210083, "learning_rate": 0.0002, "epoch": 3.9085714285714284, "step": 1710}, {"loss": 1.4317, "grad_norm": 0.6987943053245544, "learning_rate": 0.0002, "epoch": 3.9314285714285715, "step": 1720}, {"loss": 1.4704, "grad_norm": 0.7062228918075562, "learning_rate": 0.0002, "epoch": 3.954285714285714, "step": 1730}, {"loss": 1.4219, "grad_norm": 0.6769542098045349, "learning_rate": 0.0002, "epoch": 3.977142857142857, "step": 1740}, {"loss": 1.4998, "grad_norm": 0.6832144260406494, "learning_rate": 0.0002, "epoch": 4.0, "step": 1750}]}
+{"epoch": 4.998857142857143, "step": 2187, "epoch_duration": 1475.801969051361, "total_accumulated_duration": 7237.824382066727, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}, {"eval_loss": 1.8684407472610474, "eval_runtime": 111.2835, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 2.998857142857143, "step": 1312}, {"loss": 1.4486, "grad_norm": 0.5184893012046814, "learning_rate": 0.0002, "epoch": 3.0171428571428573, "step": 1320}, {"loss": 1.4082, "grad_norm": 0.5665355920791626, "learning_rate": 0.0002, "epoch": 3.04, "step": 1330}, {"loss": 1.3741, "grad_norm": 0.6601403951644897, "learning_rate": 0.0002, "epoch": 3.0628571428571427, "step": 1340}, {"loss": 1.433, "grad_norm": 0.6921621561050415, "learning_rate": 0.0002, "epoch": 3.085714285714286, "step": 1350}, {"loss": 1.4562, "grad_norm": 0.6406348943710327, "learning_rate": 0.0002, "epoch": 3.1085714285714285, "step": 1360}, {"loss": 1.3563, "grad_norm": 0.5814554691314697, "learning_rate": 0.0002, "epoch": 3.1314285714285712, "step": 1370}, {"loss": 1.4096, "grad_norm": 0.683325469493866, "learning_rate": 0.0002, "epoch": 3.1542857142857144, "step": 1380}, {"loss": 1.4106, "grad_norm": 0.6686155200004578, "learning_rate": 0.0002, "epoch": 3.177142857142857, "step": 1390}, {"loss": 1.4394, "grad_norm": 0.8159713745117188, "learning_rate": 0.0002, "epoch": 3.2, "step": 1400}, {"loss": 1.4279, "grad_norm": 0.646216094493866, "learning_rate": 0.0002, "epoch": 3.222857142857143, "step": 1410}, {"loss": 1.4232, "grad_norm": 0.7323529720306396, "learning_rate": 0.0002, "epoch": 3.2457142857142856, "step": 1420}, {"loss": 1.3891, "grad_norm": 0.689349353313446, "learning_rate": 0.0002, "epoch": 3.2685714285714287, "step": 1430}, {"loss": 1.4578, "grad_norm": 0.727894127368927, "learning_rate": 0.0002, "epoch": 3.2914285714285714, "step": 1440}, {"loss": 1.4, "grad_norm": 0.6921590566635132, "learning_rate": 0.0002, "epoch": 3.314285714285714, "step": 1450}, {"loss": 1.4272, "grad_norm": 0.6176243424415588, "learning_rate": 0.0002, "epoch": 3.337142857142857, "step": 1460}, {"loss": 1.4323, "grad_norm": 0.9006354212760925, "learning_rate": 0.0002, "epoch": 3.36, "step": 1470}, {"loss": 1.4353, "grad_norm": 0.8145929574966431, "learning_rate": 0.0002, "epoch": 3.382857142857143, "step": 1480}, {"loss": 1.3859, "grad_norm": 0.6640016436576843, "learning_rate": 0.0002, "epoch": 3.4057142857142857, "step": 1490}, {"loss": 1.387, "grad_norm": 0.7266780138015747, "learning_rate": 0.0002, "epoch": 3.4285714285714284, "step": 1500}, {"loss": 1.4108, "grad_norm": 0.9351356029510498, "learning_rate": 0.0002, "epoch": 3.4514285714285715, "step": 1510}, {"loss": 1.4656, "grad_norm": 0.675645649433136, "learning_rate": 0.0002, "epoch": 3.474285714285714, "step": 1520}, {"loss": 1.384, "grad_norm": 0.761472225189209, "learning_rate": 0.0002, "epoch": 3.4971428571428573, "step": 1530}, {"loss": 1.4968, "grad_norm": 0.6653069257736206, "learning_rate": 0.0002, "epoch": 3.52, "step": 1540}, {"loss": 1.4686, "grad_norm": 0.667412519454956, "learning_rate": 0.0002, "epoch": 3.5428571428571427, "step": 1550}, {"loss": 1.4241, "grad_norm": 0.6395593881607056, "learning_rate": 0.0002, "epoch": 3.565714285714286, "step": 1560}, {"loss": 1.4825, "grad_norm": 0.7588621377944946, "learning_rate": 0.0002, "epoch": 3.5885714285714285, "step": 1570}, {"loss": 1.4459, "grad_norm": 0.6206456422805786, "learning_rate": 0.0002, "epoch": 3.611428571428571, "step": 1580}, {"loss": 1.436, "grad_norm": 0.7591291666030884, "learning_rate": 0.0002, "epoch": 3.6342857142857143, "step": 1590}, {"loss": 1.458, "grad_norm": 0.6476313471794128, "learning_rate": 0.0002, "epoch": 3.657142857142857, "step": 1600}, {"loss": 1.4598, "grad_norm": 0.6731392741203308, "learning_rate": 0.0002, "epoch": 3.68, "step": 1610}, {"loss": 1.4225, "grad_norm": 0.725190281867981, "learning_rate": 0.0002, "epoch": 3.702857142857143, "step": 1620}, {"loss": 1.4525, "grad_norm": 0.6720049977302551, "learning_rate": 0.0002, "epoch": 3.725714285714286, "step": 1630}, {"loss": 1.429, "grad_norm": 0.6301007270812988, "learning_rate": 0.0002, "epoch": 3.7485714285714287, "step": 1640}, {"loss": 1.4166, "grad_norm": 0.715893566608429, "learning_rate": 0.0002, "epoch": 3.7714285714285714, "step": 1650}, {"loss": 1.3624, "grad_norm": 0.7539359927177429, "learning_rate": 0.0002, "epoch": 3.7942857142857145, "step": 1660}, {"loss": 1.4516, "grad_norm": 0.6658543348312378, "learning_rate": 0.0002, "epoch": 3.817142857142857, "step": 1670}, {"loss": 1.3934, "grad_norm": 0.7019526958465576, "learning_rate": 0.0002, "epoch": 3.84, "step": 1680}, {"loss": 1.4436, "grad_norm": 0.6517802476882935, "learning_rate": 0.0002, "epoch": 3.862857142857143, "step": 1690}, {"loss": 1.4968, "grad_norm": 0.7617332935333252, "learning_rate": 0.0002, "epoch": 3.8857142857142857, "step": 1700}, {"loss": 1.5145, "grad_norm": 0.6919480562210083, "learning_rate": 0.0002, "epoch": 3.9085714285714284, "step": 1710}, {"loss": 1.4317, "grad_norm": 0.6987943053245544, "learning_rate": 0.0002, "epoch": 3.9314285714285715, "step": 1720}, {"loss": 1.4704, "grad_norm": 0.7062228918075562, "learning_rate": 0.0002, "epoch": 3.954285714285714, "step": 1730}, {"loss": 1.4219, "grad_norm": 0.6769542098045349, "learning_rate": 0.0002, "epoch": 3.977142857142857, "step": 1740}, {"loss": 1.4998, "grad_norm": 0.6832144260406494, "learning_rate": 0.0002, "epoch": 4.0, "step": 1750}, {"eval_loss": 1.9474865198135376, "eval_runtime": 111.288, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.0, "step": 1750}, {"loss": 1.2251, "grad_norm": 1.064110279083252, "learning_rate": 0.0002, "epoch": 4.022857142857143, "step": 1760}, {"loss": 1.2013, "grad_norm": 0.8380683660507202, "learning_rate": 0.0002, "epoch": 4.045714285714285, "step": 1770}, {"loss": 1.2416, "grad_norm": 1.1863020658493042, "learning_rate": 0.0002, "epoch": 4.0685714285714285, "step": 1780}, {"loss": 1.2499, "grad_norm": 1.0128898620605469, "learning_rate": 0.0002, "epoch": 4.091428571428572, "step": 1790}, {"loss": 1.2043, "grad_norm": 0.9221312403678894, "learning_rate": 0.0002, "epoch": 4.114285714285714, "step": 1800}, {"loss": 1.181, "grad_norm": 1.1298727989196777, "learning_rate": 0.0002, "epoch": 4.137142857142857, "step": 1810}, {"loss": 1.1491, "grad_norm": 0.8854547739028931, "learning_rate": 0.0002, "epoch": 4.16, "step": 1820}, {"loss": 1.2156, "grad_norm": 0.8920808434486389, "learning_rate": 0.0002, "epoch": 4.182857142857143, "step": 1830}, {"loss": 1.1969, "grad_norm": 0.913244366645813, "learning_rate": 0.0002, "epoch": 4.2057142857142855, "step": 1840}, {"loss": 1.2156, "grad_norm": 0.908831000328064, "learning_rate": 0.0002, "epoch": 4.228571428571429, "step": 1850}, {"loss": 1.1653, "grad_norm": 1.0223685503005981, "learning_rate": 0.0002, "epoch": 4.251428571428572, "step": 1860}, {"loss": 1.2497, "grad_norm": 0.9771921634674072, "learning_rate": 0.0002, "epoch": 4.274285714285714, "step": 1870}, {"loss": 1.213, "grad_norm": 0.9313384890556335, "learning_rate": 0.0002, "epoch": 4.297142857142857, "step": 1880}, {"loss": 1.1723, "grad_norm": 1.0754257440567017, "learning_rate": 0.0002, "epoch": 4.32, "step": 1890}, {"loss": 1.2286, "grad_norm": 0.8904672265052795, "learning_rate": 0.0002, "epoch": 4.3428571428571425, "step": 1900}, {"loss": 1.2618, "grad_norm": 1.046527624130249, "learning_rate": 0.0002, "epoch": 4.365714285714286, "step": 1910}, {"loss": 1.2368, "grad_norm": 0.9576982855796814, "learning_rate": 0.0002, "epoch": 4.388571428571429, "step": 1920}, {"loss": 1.211, "grad_norm": 0.9278356432914734, "learning_rate": 0.0002, "epoch": 4.411428571428571, "step": 1930}, {"loss": 1.2005, "grad_norm": 1.1763030290603638, "learning_rate": 0.0002, "epoch": 4.434285714285714, "step": 1940}, {"loss": 1.1541, "grad_norm": 0.9183000326156616, "learning_rate": 0.0002, "epoch": 4.457142857142857, "step": 1950}, {"loss": 1.2257, "grad_norm": 1.050980806350708, "learning_rate": 0.0002, "epoch": 4.48, "step": 1960}, {"loss": 1.2133, "grad_norm": 0.9975392818450928, "learning_rate": 0.0002, "epoch": 4.502857142857143, "step": 1970}, {"loss": 1.2312, "grad_norm": 0.990544319152832, "learning_rate": 0.0002, "epoch": 4.525714285714286, "step": 1980}, {"loss": 1.2465, "grad_norm": 1.004794955253601, "learning_rate": 0.0002, "epoch": 4.548571428571429, "step": 1990}, {"loss": 1.2085, "grad_norm": 0.9294857978820801, "learning_rate": 0.0002, "epoch": 4.571428571428571, "step": 2000}, {"loss": 1.2874, "grad_norm": 0.93436598777771, "learning_rate": 0.0002, "epoch": 4.594285714285714, "step": 2010}, {"loss": 1.1965, "grad_norm": 0.8704655766487122, "learning_rate": 0.0002, "epoch": 4.617142857142857, "step": 2020}, {"loss": 1.204, "grad_norm": 0.9077927470207214, "learning_rate": 0.0002, "epoch": 4.64, "step": 2030}, {"loss": 1.2198, "grad_norm": 0.912987470626831, "learning_rate": 0.0002, "epoch": 4.662857142857143, "step": 2040}, {"loss": 1.2868, "grad_norm": 0.9740643501281738, "learning_rate": 0.0002, "epoch": 4.685714285714286, "step": 2050}, {"loss": 1.249, "grad_norm": 1.133357048034668, "learning_rate": 0.0002, "epoch": 4.708571428571428, "step": 2060}, {"loss": 1.1974, "grad_norm": 0.8844527006149292, "learning_rate": 0.0002, "epoch": 4.731428571428571, "step": 2070}, {"loss": 1.2481, "grad_norm": 1.0083311796188354, "learning_rate": 0.0002, "epoch": 4.7542857142857144, "step": 2080}, {"loss": 1.263, "grad_norm": 1.000447154045105, "learning_rate": 0.0002, "epoch": 4.777142857142858, "step": 2090}, {"loss": 1.2313, "grad_norm": 0.9620300531387329, "learning_rate": 0.0002, "epoch": 4.8, "step": 2100}, {"loss": 1.2659, "grad_norm": 0.9843335151672363, "learning_rate": 0.0002, "epoch": 4.822857142857143, "step": 2110}, {"loss": 1.2535, "grad_norm": 0.9906681180000305, "learning_rate": 0.0002, "epoch": 4.845714285714286, "step": 2120}, {"loss": 1.2325, "grad_norm": 0.9544073939323425, "learning_rate": 0.0002, "epoch": 4.868571428571428, "step": 2130}, {"loss": 1.284, "grad_norm": 0.9392994046211243, "learning_rate": 0.0002, "epoch": 4.8914285714285715, "step": 2140}, {"loss": 1.3075, "grad_norm": 1.104519248008728, "learning_rate": 0.0002, "epoch": 4.914285714285715, "step": 2150}, {"loss": 1.2753, "grad_norm": 0.9495956897735596, "learning_rate": 0.0002, "epoch": 4.937142857142857, "step": 2160}, {"loss": 1.2412, "grad_norm": 0.9696287512779236, "learning_rate": 0.0002, "epoch": 4.96, "step": 2170}, {"loss": 1.2354, "grad_norm": 0.9933681488037109, "learning_rate": 0.0002, "epoch": 4.982857142857143, "step": 2180}]}
+{"epoch": 6.0, "step": 2625, "epoch_duration": 1401.2988345623016, "total_accumulated_duration": 8639.123216629028, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}, {"eval_loss": 1.8684407472610474, "eval_runtime": 111.2835, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 2.998857142857143, "step": 1312}, {"loss": 1.4486, "grad_norm": 0.5184893012046814, "learning_rate": 0.0002, "epoch": 3.0171428571428573, "step": 1320}, {"loss": 1.4082, "grad_norm": 0.5665355920791626, "learning_rate": 0.0002, "epoch": 3.04, "step": 1330}, {"loss": 1.3741, "grad_norm": 0.6601403951644897, "learning_rate": 0.0002, "epoch": 3.0628571428571427, "step": 1340}, {"loss": 1.433, "grad_norm": 0.6921621561050415, "learning_rate": 0.0002, "epoch": 3.085714285714286, "step": 1350}, {"loss": 1.4562, "grad_norm": 0.6406348943710327, "learning_rate": 0.0002, "epoch": 3.1085714285714285, "step": 1360}, {"loss": 1.3563, "grad_norm": 0.5814554691314697, "learning_rate": 0.0002, "epoch": 3.1314285714285712, "step": 1370}, {"loss": 1.4096, "grad_norm": 0.683325469493866, "learning_rate": 0.0002, "epoch": 3.1542857142857144, "step": 1380}, {"loss": 1.4106, "grad_norm": 0.6686155200004578, "learning_rate": 0.0002, "epoch": 3.177142857142857, "step": 1390}, {"loss": 1.4394, "grad_norm": 0.8159713745117188, "learning_rate": 0.0002, "epoch": 3.2, "step": 1400}, {"loss": 1.4279, "grad_norm": 0.646216094493866, "learning_rate": 0.0002, "epoch": 3.222857142857143, "step": 1410}, {"loss": 1.4232, "grad_norm": 0.7323529720306396, "learning_rate": 0.0002, "epoch": 3.2457142857142856, "step": 1420}, {"loss": 1.3891, "grad_norm": 0.689349353313446, "learning_rate": 0.0002, "epoch": 3.2685714285714287, "step": 1430}, {"loss": 1.4578, "grad_norm": 0.727894127368927, "learning_rate": 0.0002, "epoch": 3.2914285714285714, "step": 1440}, {"loss": 1.4, "grad_norm": 0.6921590566635132, "learning_rate": 0.0002, "epoch": 3.314285714285714, "step": 1450}, {"loss": 1.4272, "grad_norm": 0.6176243424415588, "learning_rate": 0.0002, "epoch": 3.337142857142857, "step": 1460}, {"loss": 1.4323, "grad_norm": 0.9006354212760925, "learning_rate": 0.0002, "epoch": 3.36, "step": 1470}, {"loss": 1.4353, "grad_norm": 0.8145929574966431, "learning_rate": 0.0002, "epoch": 3.382857142857143, "step": 1480}, {"loss": 1.3859, "grad_norm": 0.6640016436576843, "learning_rate": 0.0002, "epoch": 3.4057142857142857, "step": 1490}, {"loss": 1.387, "grad_norm": 0.7266780138015747, "learning_rate": 0.0002, "epoch": 3.4285714285714284, "step": 1500}, {"loss": 1.4108, "grad_norm": 0.9351356029510498, "learning_rate": 0.0002, "epoch": 3.4514285714285715, "step": 1510}, {"loss": 1.4656, "grad_norm": 0.675645649433136, "learning_rate": 0.0002, "epoch": 3.474285714285714, "step": 1520}, {"loss": 1.384, "grad_norm": 0.761472225189209, "learning_rate": 0.0002, "epoch": 3.4971428571428573, "step": 1530}, {"loss": 1.4968, "grad_norm": 0.6653069257736206, "learning_rate": 0.0002, "epoch": 3.52, "step": 1540}, {"loss": 1.4686, "grad_norm": 0.667412519454956, "learning_rate": 0.0002, "epoch": 3.5428571428571427, "step": 1550}, {"loss": 1.4241, "grad_norm": 0.6395593881607056, "learning_rate": 0.0002, "epoch": 3.565714285714286, "step": 1560}, {"loss": 1.4825, "grad_norm": 0.7588621377944946, "learning_rate": 0.0002, "epoch": 3.5885714285714285, "step": 1570}, {"loss": 1.4459, "grad_norm": 0.6206456422805786, "learning_rate": 0.0002, "epoch": 3.611428571428571, "step": 1580}, {"loss": 1.436, "grad_norm": 0.7591291666030884, "learning_rate": 0.0002, "epoch": 3.6342857142857143, "step": 1590}, {"loss": 1.458, "grad_norm": 0.6476313471794128, "learning_rate": 0.0002, "epoch": 3.657142857142857, "step": 1600}, {"loss": 1.4598, "grad_norm": 0.6731392741203308, "learning_rate": 0.0002, "epoch": 3.68, "step": 1610}, {"loss": 1.4225, "grad_norm": 0.725190281867981, "learning_rate": 0.0002, "epoch": 3.702857142857143, "step": 1620}, {"loss": 1.4525, "grad_norm": 0.6720049977302551, "learning_rate": 0.0002, "epoch": 3.725714285714286, "step": 1630}, {"loss": 1.429, "grad_norm": 0.6301007270812988, "learning_rate": 0.0002, "epoch": 3.7485714285714287, "step": 1640}, {"loss": 1.4166, "grad_norm": 0.715893566608429, "learning_rate": 0.0002, "epoch": 3.7714285714285714, "step": 1650}, {"loss": 1.3624, "grad_norm": 0.7539359927177429, "learning_rate": 0.0002, "epoch": 3.7942857142857145, "step": 1660}, {"loss": 1.4516, "grad_norm": 0.6658543348312378, "learning_rate": 0.0002, "epoch": 3.817142857142857, "step": 1670}, {"loss": 1.3934, "grad_norm": 0.7019526958465576, "learning_rate": 0.0002, "epoch": 3.84, "step": 1680}, {"loss": 1.4436, "grad_norm": 0.6517802476882935, "learning_rate": 0.0002, "epoch": 3.862857142857143, "step": 1690}, {"loss": 1.4968, "grad_norm": 0.7617332935333252, "learning_rate": 0.0002, "epoch": 3.8857142857142857, "step": 1700}, {"loss": 1.5145, "grad_norm": 0.6919480562210083, "learning_rate": 0.0002, "epoch": 3.9085714285714284, "step": 1710}, {"loss": 1.4317, "grad_norm": 0.6987943053245544, "learning_rate": 0.0002, "epoch": 3.9314285714285715, "step": 1720}, {"loss": 1.4704, "grad_norm": 0.7062228918075562, "learning_rate": 0.0002, "epoch": 3.954285714285714, "step": 1730}, {"loss": 1.4219, "grad_norm": 0.6769542098045349, "learning_rate": 0.0002, "epoch": 3.977142857142857, "step": 1740}, {"loss": 1.4998, "grad_norm": 0.6832144260406494, "learning_rate": 0.0002, "epoch": 4.0, "step": 1750}, {"eval_loss": 1.9474865198135376, "eval_runtime": 111.288, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.0, "step": 1750}, {"loss": 1.2251, "grad_norm": 1.064110279083252, "learning_rate": 0.0002, "epoch": 4.022857142857143, "step": 1760}, {"loss": 1.2013, "grad_norm": 0.8380683660507202, "learning_rate": 0.0002, "epoch": 4.045714285714285, "step": 1770}, {"loss": 1.2416, "grad_norm": 1.1863020658493042, "learning_rate": 0.0002, "epoch": 4.0685714285714285, "step": 1780}, {"loss": 1.2499, "grad_norm": 1.0128898620605469, "learning_rate": 0.0002, "epoch": 4.091428571428572, "step": 1790}, {"loss": 1.2043, "grad_norm": 0.9221312403678894, "learning_rate": 0.0002, "epoch": 4.114285714285714, "step": 1800}, {"loss": 1.181, "grad_norm": 1.1298727989196777, "learning_rate": 0.0002, "epoch": 4.137142857142857, "step": 1810}, {"loss": 1.1491, "grad_norm": 0.8854547739028931, "learning_rate": 0.0002, "epoch": 4.16, "step": 1820}, {"loss": 1.2156, "grad_norm": 0.8920808434486389, "learning_rate": 0.0002, "epoch": 4.182857142857143, "step": 1830}, {"loss": 1.1969, "grad_norm": 0.913244366645813, "learning_rate": 0.0002, "epoch": 4.2057142857142855, "step": 1840}, {"loss": 1.2156, "grad_norm": 0.908831000328064, "learning_rate": 0.0002, "epoch": 4.228571428571429, "step": 1850}, {"loss": 1.1653, "grad_norm": 1.0223685503005981, "learning_rate": 0.0002, "epoch": 4.251428571428572, "step": 1860}, {"loss": 1.2497, "grad_norm": 0.9771921634674072, "learning_rate": 0.0002, "epoch": 4.274285714285714, "step": 1870}, {"loss": 1.213, "grad_norm": 0.9313384890556335, "learning_rate": 0.0002, "epoch": 4.297142857142857, "step": 1880}, {"loss": 1.1723, "grad_norm": 1.0754257440567017, "learning_rate": 0.0002, "epoch": 4.32, "step": 1890}, {"loss": 1.2286, "grad_norm": 0.8904672265052795, "learning_rate": 0.0002, "epoch": 4.3428571428571425, "step": 1900}, {"loss": 1.2618, "grad_norm": 1.046527624130249, "learning_rate": 0.0002, "epoch": 4.365714285714286, "step": 1910}, {"loss": 1.2368, "grad_norm": 0.9576982855796814, "learning_rate": 0.0002, "epoch": 4.388571428571429, "step": 1920}, {"loss": 1.211, "grad_norm": 0.9278356432914734, "learning_rate": 0.0002, "epoch": 4.411428571428571, "step": 1930}, {"loss": 1.2005, "grad_norm": 1.1763030290603638, "learning_rate": 0.0002, "epoch": 4.434285714285714, "step": 1940}, {"loss": 1.1541, "grad_norm": 0.9183000326156616, "learning_rate": 0.0002, "epoch": 4.457142857142857, "step": 1950}, {"loss": 1.2257, "grad_norm": 1.050980806350708, "learning_rate": 0.0002, "epoch": 4.48, "step": 1960}, {"loss": 1.2133, "grad_norm": 0.9975392818450928, "learning_rate": 0.0002, "epoch": 4.502857142857143, "step": 1970}, {"loss": 1.2312, "grad_norm": 0.990544319152832, "learning_rate": 0.0002, "epoch": 4.525714285714286, "step": 1980}, {"loss": 1.2465, "grad_norm": 1.004794955253601, "learning_rate": 0.0002, "epoch": 4.548571428571429, "step": 1990}, {"loss": 1.2085, "grad_norm": 0.9294857978820801, "learning_rate": 0.0002, "epoch": 4.571428571428571, "step": 2000}, {"loss": 1.2874, "grad_norm": 0.93436598777771, "learning_rate": 0.0002, "epoch": 4.594285714285714, "step": 2010}, {"loss": 1.1965, "grad_norm": 0.8704655766487122, "learning_rate": 0.0002, "epoch": 4.617142857142857, "step": 2020}, {"loss": 1.204, "grad_norm": 0.9077927470207214, "learning_rate": 0.0002, "epoch": 4.64, "step": 2030}, {"loss": 1.2198, "grad_norm": 0.912987470626831, "learning_rate": 0.0002, "epoch": 4.662857142857143, "step": 2040}, {"loss": 1.2868, "grad_norm": 0.9740643501281738, "learning_rate": 0.0002, "epoch": 4.685714285714286, "step": 2050}, {"loss": 1.249, "grad_norm": 1.133357048034668, "learning_rate": 0.0002, "epoch": 4.708571428571428, "step": 2060}, {"loss": 1.1974, "grad_norm": 0.8844527006149292, "learning_rate": 0.0002, "epoch": 4.731428571428571, "step": 2070}, {"loss": 1.2481, "grad_norm": 1.0083311796188354, "learning_rate": 0.0002, "epoch": 4.7542857142857144, "step": 2080}, {"loss": 1.263, "grad_norm": 1.000447154045105, "learning_rate": 0.0002, "epoch": 4.777142857142858, "step": 2090}, {"loss": 1.2313, "grad_norm": 0.9620300531387329, "learning_rate": 0.0002, "epoch": 4.8, "step": 2100}, {"loss": 1.2659, "grad_norm": 0.9843335151672363, "learning_rate": 0.0002, "epoch": 4.822857142857143, "step": 2110}, {"loss": 1.2535, "grad_norm": 0.9906681180000305, "learning_rate": 0.0002, "epoch": 4.845714285714286, "step": 2120}, {"loss": 1.2325, "grad_norm": 0.9544073939323425, "learning_rate": 0.0002, "epoch": 4.868571428571428, "step": 2130}, {"loss": 1.284, "grad_norm": 0.9392994046211243, "learning_rate": 0.0002, "epoch": 4.8914285714285715, "step": 2140}, {"loss": 1.3075, "grad_norm": 1.104519248008728, "learning_rate": 0.0002, "epoch": 4.914285714285715, "step": 2150}, {"loss": 1.2753, "grad_norm": 0.9495956897735596, "learning_rate": 0.0002, "epoch": 4.937142857142857, "step": 2160}, {"loss": 1.2412, "grad_norm": 0.9696287512779236, "learning_rate": 0.0002, "epoch": 4.96, "step": 2170}, {"loss": 1.2354, "grad_norm": 0.9933681488037109, "learning_rate": 0.0002, "epoch": 4.982857142857143, "step": 2180}, {"eval_loss": 2.099808692932129, "eval_runtime": 111.2808, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.998857142857143, "step": 2187}, {"loss": 1.2183, "grad_norm": 0.9482853412628174, "learning_rate": 0.0002, "epoch": 5.005714285714285, "step": 2190}, {"loss": 0.9898, "grad_norm": 1.6689555644989014, "learning_rate": 0.0002, "epoch": 5.0285714285714285, "step": 2200}, {"loss": 0.9741, "grad_norm": 1.2019699811935425, "learning_rate": 0.0002, "epoch": 5.051428571428572, "step": 2210}, {"loss": 0.9737, "grad_norm": 1.535780429840088, "learning_rate": 0.0002, "epoch": 5.074285714285715, "step": 2220}, {"loss": 0.9494, "grad_norm": 1.2061309814453125, "learning_rate": 0.0002, "epoch": 5.097142857142857, "step": 2230}, {"loss": 0.9316, "grad_norm": 1.1898778676986694, "learning_rate": 0.0002, "epoch": 5.12, "step": 2240}, {"loss": 1.002, "grad_norm": 1.158898949623108, "learning_rate": 0.0002, "epoch": 5.142857142857143, "step": 2250}, {"loss": 0.9715, "grad_norm": 1.370749592781067, "learning_rate": 0.0002, "epoch": 5.1657142857142855, "step": 2260}, {"loss": 0.9365, "grad_norm": 1.314120888710022, "learning_rate": 0.0002, "epoch": 5.188571428571429, "step": 2270}, {"loss": 1.0316, "grad_norm": 1.2184966802597046, "learning_rate": 0.0002, "epoch": 5.211428571428572, "step": 2280}, {"loss": 0.9407, "grad_norm": 1.4833279848098755, "learning_rate": 0.0002, "epoch": 5.234285714285714, "step": 2290}, {"loss": 0.9635, "grad_norm": 1.3348219394683838, "learning_rate": 0.0002, "epoch": 5.257142857142857, "step": 2300}, {"loss": 1.0294, "grad_norm": 1.4166619777679443, "learning_rate": 0.0002, "epoch": 5.28, "step": 2310}, {"loss": 0.9818, "grad_norm": 1.4539530277252197, "learning_rate": 0.0002, "epoch": 5.3028571428571425, "step": 2320}, {"loss": 1.0165, "grad_norm": 1.4642518758773804, "learning_rate": 0.0002, "epoch": 5.325714285714286, "step": 2330}, {"loss": 1.0081, "grad_norm": 1.3938848972320557, "learning_rate": 0.0002, "epoch": 5.348571428571429, "step": 2340}, {"loss": 1.03, "grad_norm": 1.1147894859313965, "learning_rate": 0.0002, "epoch": 5.371428571428572, "step": 2350}, {"loss": 0.9975, "grad_norm": 1.3465309143066406, "learning_rate": 0.0002, "epoch": 5.394285714285714, "step": 2360}, {"loss": 1.0138, "grad_norm": 1.4788566827774048, "learning_rate": 0.0002, "epoch": 5.417142857142857, "step": 2370}, {"loss": 0.9896, "grad_norm": 1.3808705806732178, "learning_rate": 0.0002, "epoch": 5.44, "step": 2380}, {"loss": 1.0279, "grad_norm": 1.2336329221725464, "learning_rate": 0.0002, "epoch": 5.462857142857143, "step": 2390}, {"loss": 0.9763, "grad_norm": 1.5445678234100342, "learning_rate": 0.0002, "epoch": 5.485714285714286, "step": 2400}, {"loss": 0.9534, "grad_norm": 1.107488989830017, "learning_rate": 0.0002, "epoch": 5.508571428571429, "step": 2410}, {"loss": 1.0036, "grad_norm": 1.39687979221344, "learning_rate": 0.0002, "epoch": 5.531428571428571, "step": 2420}, {"loss": 0.9959, "grad_norm": 1.3905695676803589, "learning_rate": 0.0002, "epoch": 5.554285714285714, "step": 2430}, {"loss": 0.9912, "grad_norm": 1.3772821426391602, "learning_rate": 0.0002, "epoch": 5.577142857142857, "step": 2440}, {"loss": 0.9825, "grad_norm": 1.1661899089813232, "learning_rate": 0.0002, "epoch": 5.6, "step": 2450}, {"loss": 1.0003, "grad_norm": 1.2730463743209839, "learning_rate": 0.0002, "epoch": 5.622857142857143, "step": 2460}, {"loss": 1.0433, "grad_norm": 1.2251193523406982, "learning_rate": 0.0002, "epoch": 5.645714285714286, "step": 2470}, {"loss": 1.079, "grad_norm": 1.5454859733581543, "learning_rate": 0.0002, "epoch": 5.668571428571429, "step": 2480}, {"loss": 1.0414, "grad_norm": 1.5405735969543457, "learning_rate": 0.0002, "epoch": 5.691428571428571, "step": 2490}, {"loss": 1.0353, "grad_norm": 1.2555434703826904, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2500}, {"loss": 1.0019, "grad_norm": 1.3323487043380737, "learning_rate": 0.0002, "epoch": 5.737142857142857, "step": 2510}, {"loss": 1.051, "grad_norm": 1.3106356859207153, "learning_rate": 0.0002, "epoch": 5.76, "step": 2520}, {"loss": 1.0248, "grad_norm": 1.4832439422607422, "learning_rate": 0.0002, "epoch": 5.782857142857143, "step": 2530}, {"loss": 1.0643, "grad_norm": 1.1336562633514404, "learning_rate": 0.0002, "epoch": 5.805714285714286, "step": 2540}, {"loss": 1.0446, "grad_norm": 1.2434223890304565, "learning_rate": 0.0002, "epoch": 5.828571428571428, "step": 2550}, {"loss": 1.0467, "grad_norm": 1.2825450897216797, "learning_rate": 0.0002, "epoch": 5.851428571428571, "step": 2560}, {"loss": 1.0642, "grad_norm": 1.4373180866241455, "learning_rate": 0.0002, "epoch": 5.8742857142857146, "step": 2570}, {"loss": 1.0814, "grad_norm": 1.435015320777893, "learning_rate": 0.0002, "epoch": 5.897142857142857, "step": 2580}, {"loss": 1.0272, "grad_norm": 1.4075653553009033, "learning_rate": 0.0002, "epoch": 5.92, "step": 2590}, {"loss": 1.0703, "grad_norm": 1.319630742073059, "learning_rate": 0.0002, "epoch": 5.942857142857143, "step": 2600}, {"loss": 1.0375, "grad_norm": 1.278330683708191, "learning_rate": 0.0002, "epoch": 5.965714285714286, "step": 2610}, {"loss": 1.0766, "grad_norm": 1.258158564567566, "learning_rate": 0.0002, "epoch": 5.988571428571428, "step": 2620}]}
+{"epoch": 6.998857142857143, "step": 3062, "epoch_duration": 774.590334892273, "total_accumulated_duration": 9413.713551521301, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}, {"eval_loss": 1.8684407472610474, "eval_runtime": 111.2835, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 2.998857142857143, "step": 1312}, {"loss": 1.4486, "grad_norm": 0.5184893012046814, "learning_rate": 0.0002, "epoch": 3.0171428571428573, "step": 1320}, {"loss": 1.4082, "grad_norm": 0.5665355920791626, "learning_rate": 0.0002, "epoch": 3.04, "step": 1330}, {"loss": 1.3741, "grad_norm": 0.6601403951644897, "learning_rate": 0.0002, "epoch": 3.0628571428571427, "step": 1340}, {"loss": 1.433, "grad_norm": 0.6921621561050415, "learning_rate": 0.0002, "epoch": 3.085714285714286, "step": 1350}, {"loss": 1.4562, "grad_norm": 0.6406348943710327, "learning_rate": 0.0002, "epoch": 3.1085714285714285, "step": 1360}, {"loss": 1.3563, "grad_norm": 0.5814554691314697, "learning_rate": 0.0002, "epoch": 3.1314285714285712, "step": 1370}, {"loss": 1.4096, "grad_norm": 0.683325469493866, "learning_rate": 0.0002, "epoch": 3.1542857142857144, "step": 1380}, {"loss": 1.4106, "grad_norm": 0.6686155200004578, "learning_rate": 0.0002, "epoch": 3.177142857142857, "step": 1390}, {"loss": 1.4394, "grad_norm": 0.8159713745117188, "learning_rate": 0.0002, "epoch": 3.2, "step": 1400}, {"loss": 1.4279, "grad_norm": 0.646216094493866, "learning_rate": 0.0002, "epoch": 3.222857142857143, "step": 1410}, {"loss": 1.4232, "grad_norm": 0.7323529720306396, "learning_rate": 0.0002, "epoch": 3.2457142857142856, "step": 1420}, {"loss": 1.3891, "grad_norm": 0.689349353313446, "learning_rate": 0.0002, "epoch": 3.2685714285714287, "step": 1430}, {"loss": 1.4578, "grad_norm": 0.727894127368927, "learning_rate": 0.0002, "epoch": 3.2914285714285714, "step": 1440}, {"loss": 1.4, "grad_norm": 0.6921590566635132, "learning_rate": 0.0002, "epoch": 3.314285714285714, "step": 1450}, {"loss": 1.4272, "grad_norm": 0.6176243424415588, "learning_rate": 0.0002, "epoch": 3.337142857142857, "step": 1460}, {"loss": 1.4323, "grad_norm": 0.9006354212760925, "learning_rate": 0.0002, "epoch": 3.36, "step": 1470}, {"loss": 1.4353, "grad_norm": 0.8145929574966431, "learning_rate": 0.0002, "epoch": 3.382857142857143, "step": 1480}, {"loss": 1.3859, "grad_norm": 0.6640016436576843, "learning_rate": 0.0002, "epoch": 3.4057142857142857, "step": 1490}, {"loss": 1.387, "grad_norm": 0.7266780138015747, "learning_rate": 0.0002, "epoch": 3.4285714285714284, "step": 1500}, {"loss": 1.4108, "grad_norm": 0.9351356029510498, "learning_rate": 0.0002, "epoch": 3.4514285714285715, "step": 1510}, {"loss": 1.4656, "grad_norm": 0.675645649433136, "learning_rate": 0.0002, "epoch": 3.474285714285714, "step": 1520}, {"loss": 1.384, "grad_norm": 0.761472225189209, "learning_rate": 0.0002, "epoch": 3.4971428571428573, "step": 1530}, {"loss": 1.4968, "grad_norm": 0.6653069257736206, "learning_rate": 0.0002, "epoch": 3.52, "step": 1540}, {"loss": 1.4686, "grad_norm": 0.667412519454956, "learning_rate": 0.0002, "epoch": 3.5428571428571427, "step": 1550}, {"loss": 1.4241, "grad_norm": 0.6395593881607056, "learning_rate": 0.0002, "epoch": 3.565714285714286, "step": 1560}, {"loss": 1.4825, "grad_norm": 0.7588621377944946, "learning_rate": 0.0002, "epoch": 3.5885714285714285, "step": 1570}, {"loss": 1.4459, "grad_norm": 0.6206456422805786, "learning_rate": 0.0002, "epoch": 3.611428571428571, "step": 1580}, {"loss": 1.436, "grad_norm": 0.7591291666030884, "learning_rate": 0.0002, "epoch": 3.6342857142857143, "step": 1590}, {"loss": 1.458, "grad_norm": 0.6476313471794128, "learning_rate": 0.0002, "epoch": 3.657142857142857, "step": 1600}, {"loss": 1.4598, "grad_norm": 0.6731392741203308, "learning_rate": 0.0002, "epoch": 3.68, "step": 1610}, {"loss": 1.4225, "grad_norm": 0.725190281867981, "learning_rate": 0.0002, "epoch": 3.702857142857143, "step": 1620}, {"loss": 1.4525, "grad_norm": 0.6720049977302551, "learning_rate": 0.0002, "epoch": 3.725714285714286, "step": 1630}, {"loss": 1.429, "grad_norm": 0.6301007270812988, "learning_rate": 0.0002, "epoch": 3.7485714285714287, "step": 1640}, {"loss": 1.4166, "grad_norm": 0.715893566608429, "learning_rate": 0.0002, "epoch": 3.7714285714285714, "step": 1650}, {"loss": 1.3624, "grad_norm": 0.7539359927177429, "learning_rate": 0.0002, "epoch": 3.7942857142857145, "step": 1660}, {"loss": 1.4516, "grad_norm": 0.6658543348312378, "learning_rate": 0.0002, "epoch": 3.817142857142857, "step": 1670}, {"loss": 1.3934, "grad_norm": 0.7019526958465576, "learning_rate": 0.0002, "epoch": 3.84, "step": 1680}, {"loss": 1.4436, "grad_norm": 0.6517802476882935, "learning_rate": 0.0002, "epoch": 3.862857142857143, "step": 1690}, {"loss": 1.4968, "grad_norm": 0.7617332935333252, "learning_rate": 0.0002, "epoch": 3.8857142857142857, "step": 1700}, {"loss": 1.5145, "grad_norm": 0.6919480562210083, "learning_rate": 0.0002, "epoch": 3.9085714285714284, "step": 1710}, {"loss": 1.4317, "grad_norm": 0.6987943053245544, "learning_rate": 0.0002, "epoch": 3.9314285714285715, "step": 1720}, {"loss": 1.4704, "grad_norm": 0.7062228918075562, "learning_rate": 0.0002, "epoch": 3.954285714285714, "step": 1730}, {"loss": 1.4219, "grad_norm": 0.6769542098045349, "learning_rate": 0.0002, "epoch": 3.977142857142857, "step": 1740}, {"loss": 1.4998, "grad_norm": 0.6832144260406494, "learning_rate": 0.0002, "epoch": 4.0, "step": 1750}, {"eval_loss": 1.9474865198135376, "eval_runtime": 111.288, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.0, "step": 1750}, {"loss": 1.2251, "grad_norm": 1.064110279083252, "learning_rate": 0.0002, "epoch": 4.022857142857143, "step": 1760}, {"loss": 1.2013, "grad_norm": 0.8380683660507202, "learning_rate": 0.0002, "epoch": 4.045714285714285, "step": 1770}, {"loss": 1.2416, "grad_norm": 1.1863020658493042, "learning_rate": 0.0002, "epoch": 4.0685714285714285, "step": 1780}, {"loss": 1.2499, "grad_norm": 1.0128898620605469, "learning_rate": 0.0002, "epoch": 4.091428571428572, "step": 1790}, {"loss": 1.2043, "grad_norm": 0.9221312403678894, "learning_rate": 0.0002, "epoch": 4.114285714285714, "step": 1800}, {"loss": 1.181, "grad_norm": 1.1298727989196777, "learning_rate": 0.0002, "epoch": 4.137142857142857, "step": 1810}, {"loss": 1.1491, "grad_norm": 0.8854547739028931, "learning_rate": 0.0002, "epoch": 4.16, "step": 1820}, {"loss": 1.2156, "grad_norm": 0.8920808434486389, "learning_rate": 0.0002, "epoch": 4.182857142857143, "step": 1830}, {"loss": 1.1969, "grad_norm": 0.913244366645813, "learning_rate": 0.0002, "epoch": 4.2057142857142855, "step": 1840}, {"loss": 1.2156, "grad_norm": 0.908831000328064, "learning_rate": 0.0002, "epoch": 4.228571428571429, "step": 1850}, {"loss": 1.1653, "grad_norm": 1.0223685503005981, "learning_rate": 0.0002, "epoch": 4.251428571428572, "step": 1860}, {"loss": 1.2497, "grad_norm": 0.9771921634674072, "learning_rate": 0.0002, "epoch": 4.274285714285714, "step": 1870}, {"loss": 1.213, "grad_norm": 0.9313384890556335, "learning_rate": 0.0002, "epoch": 4.297142857142857, "step": 1880}, {"loss": 1.1723, "grad_norm": 1.0754257440567017, "learning_rate": 0.0002, "epoch": 4.32, "step": 1890}, {"loss": 1.2286, "grad_norm": 0.8904672265052795, "learning_rate": 0.0002, "epoch": 4.3428571428571425, "step": 1900}, {"loss": 1.2618, "grad_norm": 1.046527624130249, "learning_rate": 0.0002, "epoch": 4.365714285714286, "step": 1910}, {"loss": 1.2368, "grad_norm": 0.9576982855796814, "learning_rate": 0.0002, "epoch": 4.388571428571429, "step": 1920}, {"loss": 1.211, "grad_norm": 0.9278356432914734, "learning_rate": 0.0002, "epoch": 4.411428571428571, "step": 1930}, {"loss": 1.2005, "grad_norm": 1.1763030290603638, "learning_rate": 0.0002, "epoch": 4.434285714285714, "step": 1940}, {"loss": 1.1541, "grad_norm": 0.9183000326156616, "learning_rate": 0.0002, "epoch": 4.457142857142857, "step": 1950}, {"loss": 1.2257, "grad_norm": 1.050980806350708, "learning_rate": 0.0002, "epoch": 4.48, "step": 1960}, {"loss": 1.2133, "grad_norm": 0.9975392818450928, "learning_rate": 0.0002, "epoch": 4.502857142857143, "step": 1970}, {"loss": 1.2312, "grad_norm": 0.990544319152832, "learning_rate": 0.0002, "epoch": 4.525714285714286, "step": 1980}, {"loss": 1.2465, "grad_norm": 1.004794955253601, "learning_rate": 0.0002, "epoch": 4.548571428571429, "step": 1990}, {"loss": 1.2085, "grad_norm": 0.9294857978820801, "learning_rate": 0.0002, "epoch": 4.571428571428571, "step": 2000}, {"loss": 1.2874, "grad_norm": 0.93436598777771, "learning_rate": 0.0002, "epoch": 4.594285714285714, "step": 2010}, {"loss": 1.1965, "grad_norm": 0.8704655766487122, "learning_rate": 0.0002, "epoch": 4.617142857142857, "step": 2020}, {"loss": 1.204, "grad_norm": 0.9077927470207214, "learning_rate": 0.0002, "epoch": 4.64, "step": 2030}, {"loss": 1.2198, "grad_norm": 0.912987470626831, "learning_rate": 0.0002, "epoch": 4.662857142857143, "step": 2040}, {"loss": 1.2868, "grad_norm": 0.9740643501281738, "learning_rate": 0.0002, "epoch": 4.685714285714286, "step": 2050}, {"loss": 1.249, "grad_norm": 1.133357048034668, "learning_rate": 0.0002, "epoch": 4.708571428571428, "step": 2060}, {"loss": 1.1974, "grad_norm": 0.8844527006149292, "learning_rate": 0.0002, "epoch": 4.731428571428571, "step": 2070}, {"loss": 1.2481, "grad_norm": 1.0083311796188354, "learning_rate": 0.0002, "epoch": 4.7542857142857144, "step": 2080}, {"loss": 1.263, "grad_norm": 1.000447154045105, "learning_rate": 0.0002, "epoch": 4.777142857142858, "step": 2090}, {"loss": 1.2313, "grad_norm": 0.9620300531387329, "learning_rate": 0.0002, "epoch": 4.8, "step": 2100}, {"loss": 1.2659, "grad_norm": 0.9843335151672363, "learning_rate": 0.0002, "epoch": 4.822857142857143, "step": 2110}, {"loss": 1.2535, "grad_norm": 0.9906681180000305, "learning_rate": 0.0002, "epoch": 4.845714285714286, "step": 2120}, {"loss": 1.2325, "grad_norm": 0.9544073939323425, "learning_rate": 0.0002, "epoch": 4.868571428571428, "step": 2130}, {"loss": 1.284, "grad_norm": 0.9392994046211243, "learning_rate": 0.0002, "epoch": 4.8914285714285715, "step": 2140}, {"loss": 1.3075, "grad_norm": 1.104519248008728, "learning_rate": 0.0002, "epoch": 4.914285714285715, "step": 2150}, {"loss": 1.2753, "grad_norm": 0.9495956897735596, "learning_rate": 0.0002, "epoch": 4.937142857142857, "step": 2160}, {"loss": 1.2412, "grad_norm": 0.9696287512779236, "learning_rate": 0.0002, "epoch": 4.96, "step": 2170}, {"loss": 1.2354, "grad_norm": 0.9933681488037109, "learning_rate": 0.0002, "epoch": 4.982857142857143, "step": 2180}, {"eval_loss": 2.099808692932129, "eval_runtime": 111.2808, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.998857142857143, "step": 2187}, {"loss": 1.2183, "grad_norm": 0.9482853412628174, "learning_rate": 0.0002, "epoch": 5.005714285714285, "step": 2190}, {"loss": 0.9898, "grad_norm": 1.6689555644989014, "learning_rate": 0.0002, "epoch": 5.0285714285714285, "step": 2200}, {"loss": 0.9741, "grad_norm": 1.2019699811935425, "learning_rate": 0.0002, "epoch": 5.051428571428572, "step": 2210}, {"loss": 0.9737, "grad_norm": 1.535780429840088, "learning_rate": 0.0002, "epoch": 5.074285714285715, "step": 2220}, {"loss": 0.9494, "grad_norm": 1.2061309814453125, "learning_rate": 0.0002, "epoch": 5.097142857142857, "step": 2230}, {"loss": 0.9316, "grad_norm": 1.1898778676986694, "learning_rate": 0.0002, "epoch": 5.12, "step": 2240}, {"loss": 1.002, "grad_norm": 1.158898949623108, "learning_rate": 0.0002, "epoch": 5.142857142857143, "step": 2250}, {"loss": 0.9715, "grad_norm": 1.370749592781067, "learning_rate": 0.0002, "epoch": 5.1657142857142855, "step": 2260}, {"loss": 0.9365, "grad_norm": 1.314120888710022, "learning_rate": 0.0002, "epoch": 5.188571428571429, "step": 2270}, {"loss": 1.0316, "grad_norm": 1.2184966802597046, "learning_rate": 0.0002, "epoch": 5.211428571428572, "step": 2280}, {"loss": 0.9407, "grad_norm": 1.4833279848098755, "learning_rate": 0.0002, "epoch": 5.234285714285714, "step": 2290}, {"loss": 0.9635, "grad_norm": 1.3348219394683838, "learning_rate": 0.0002, "epoch": 5.257142857142857, "step": 2300}, {"loss": 1.0294, "grad_norm": 1.4166619777679443, "learning_rate": 0.0002, "epoch": 5.28, "step": 2310}, {"loss": 0.9818, "grad_norm": 1.4539530277252197, "learning_rate": 0.0002, "epoch": 5.3028571428571425, "step": 2320}, {"loss": 1.0165, "grad_norm": 1.4642518758773804, "learning_rate": 0.0002, "epoch": 5.325714285714286, "step": 2330}, {"loss": 1.0081, "grad_norm": 1.3938848972320557, "learning_rate": 0.0002, "epoch": 5.348571428571429, "step": 2340}, {"loss": 1.03, "grad_norm": 1.1147894859313965, "learning_rate": 0.0002, "epoch": 5.371428571428572, "step": 2350}, {"loss": 0.9975, "grad_norm": 1.3465309143066406, "learning_rate": 0.0002, "epoch": 5.394285714285714, "step": 2360}, {"loss": 1.0138, "grad_norm": 1.4788566827774048, "learning_rate": 0.0002, "epoch": 5.417142857142857, "step": 2370}, {"loss": 0.9896, "grad_norm": 1.3808705806732178, "learning_rate": 0.0002, "epoch": 5.44, "step": 2380}, {"loss": 1.0279, "grad_norm": 1.2336329221725464, "learning_rate": 0.0002, "epoch": 5.462857142857143, "step": 2390}, {"loss": 0.9763, "grad_norm": 1.5445678234100342, "learning_rate": 0.0002, "epoch": 5.485714285714286, "step": 2400}, {"loss": 0.9534, "grad_norm": 1.107488989830017, "learning_rate": 0.0002, "epoch": 5.508571428571429, "step": 2410}, {"loss": 1.0036, "grad_norm": 1.39687979221344, "learning_rate": 0.0002, "epoch": 5.531428571428571, "step": 2420}, {"loss": 0.9959, "grad_norm": 1.3905695676803589, "learning_rate": 0.0002, "epoch": 5.554285714285714, "step": 2430}, {"loss": 0.9912, "grad_norm": 1.3772821426391602, "learning_rate": 0.0002, "epoch": 5.577142857142857, "step": 2440}, {"loss": 0.9825, "grad_norm": 1.1661899089813232, "learning_rate": 0.0002, "epoch": 5.6, "step": 2450}, {"loss": 1.0003, "grad_norm": 1.2730463743209839, "learning_rate": 0.0002, "epoch": 5.622857142857143, "step": 2460}, {"loss": 1.0433, "grad_norm": 1.2251193523406982, "learning_rate": 0.0002, "epoch": 5.645714285714286, "step": 2470}, {"loss": 1.079, "grad_norm": 1.5454859733581543, "learning_rate": 0.0002, "epoch": 5.668571428571429, "step": 2480}, {"loss": 1.0414, "grad_norm": 1.5405735969543457, "learning_rate": 0.0002, "epoch": 5.691428571428571, "step": 2490}, {"loss": 1.0353, "grad_norm": 1.2555434703826904, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2500}, {"loss": 1.0019, "grad_norm": 1.3323487043380737, "learning_rate": 0.0002, "epoch": 5.737142857142857, "step": 2510}, {"loss": 1.051, "grad_norm": 1.3106356859207153, "learning_rate": 0.0002, "epoch": 5.76, "step": 2520}, {"loss": 1.0248, "grad_norm": 1.4832439422607422, "learning_rate": 0.0002, "epoch": 5.782857142857143, "step": 2530}, {"loss": 1.0643, "grad_norm": 1.1336562633514404, "learning_rate": 0.0002, "epoch": 5.805714285714286, "step": 2540}, {"loss": 1.0446, "grad_norm": 1.2434223890304565, "learning_rate": 0.0002, "epoch": 5.828571428571428, "step": 2550}, {"loss": 1.0467, "grad_norm": 1.2825450897216797, "learning_rate": 0.0002, "epoch": 5.851428571428571, "step": 2560}, {"loss": 1.0642, "grad_norm": 1.4373180866241455, "learning_rate": 0.0002, "epoch": 5.8742857142857146, "step": 2570}, {"loss": 1.0814, "grad_norm": 1.435015320777893, "learning_rate": 0.0002, "epoch": 5.897142857142857, "step": 2580}, {"loss": 1.0272, "grad_norm": 1.4075653553009033, "learning_rate": 0.0002, "epoch": 5.92, "step": 2590}, {"loss": 1.0703, "grad_norm": 1.319630742073059, "learning_rate": 0.0002, "epoch": 5.942857142857143, "step": 2600}, {"loss": 1.0375, "grad_norm": 1.278330683708191, "learning_rate": 0.0002, "epoch": 5.965714285714286, "step": 2610}, {"loss": 1.0766, "grad_norm": 1.258158564567566, "learning_rate": 0.0002, "epoch": 5.988571428571428, "step": 2620}, {"eval_loss": 2.3689301013946533, "eval_runtime": 53.9067, "eval_samples_per_second": 9.405, "eval_steps_per_second": 1.187, "epoch": 6.0, "step": 2625}, {"loss": 0.9142, "grad_norm": 1.3128368854522705, "learning_rate": 0.0002, "epoch": 6.011428571428572, "step": 2630}, {"loss": 0.7716, "grad_norm": 1.4280474185943604, "learning_rate": 0.0002, "epoch": 6.034285714285715, "step": 2640}, {"loss": 0.7776, "grad_norm": 1.5061450004577637, "learning_rate": 0.0002, "epoch": 6.057142857142857, "step": 2650}, {"loss": 0.7707, "grad_norm": 1.6013342142105103, "learning_rate": 0.0002, "epoch": 6.08, "step": 2660}, {"loss": 0.7543, "grad_norm": 2.0107381343841553, "learning_rate": 0.0002, "epoch": 6.102857142857143, "step": 2670}, {"loss": 0.747, "grad_norm": 1.5010124444961548, "learning_rate": 0.0002, "epoch": 6.1257142857142854, "step": 2680}, {"loss": 0.7501, "grad_norm": 1.5222150087356567, "learning_rate": 0.0002, "epoch": 6.148571428571429, "step": 2690}, {"loss": 0.7712, "grad_norm": 1.5413103103637695, "learning_rate": 0.0002, "epoch": 6.171428571428572, "step": 2700}, {"loss": 0.7, "grad_norm": 1.527140736579895, "learning_rate": 0.0002, "epoch": 6.194285714285714, "step": 2710}, {"loss": 0.7539, "grad_norm": 1.9386590719223022, "learning_rate": 0.0002, "epoch": 6.217142857142857, "step": 2720}, {"loss": 0.7586, "grad_norm": 1.8115214109420776, "learning_rate": 0.0002, "epoch": 6.24, "step": 2730}, {"loss": 0.7426, "grad_norm": 1.6221802234649658, "learning_rate": 0.0002, "epoch": 6.2628571428571425, "step": 2740}, {"loss": 0.8002, "grad_norm": 1.6698768138885498, "learning_rate": 0.0002, "epoch": 6.285714285714286, "step": 2750}, {"loss": 0.7293, "grad_norm": 1.7960610389709473, "learning_rate": 0.0002, "epoch": 6.308571428571429, "step": 2760}, {"loss": 0.7405, "grad_norm": 1.32172429561615, "learning_rate": 0.0002, "epoch": 6.331428571428571, "step": 2770}, {"loss": 0.7198, "grad_norm": 1.7468090057373047, "learning_rate": 0.0002, "epoch": 6.354285714285714, "step": 2780}, {"loss": 0.76, "grad_norm": 1.6777397394180298, "learning_rate": 0.0002, "epoch": 6.377142857142857, "step": 2790}, {"loss": 0.7879, "grad_norm": 1.6200671195983887, "learning_rate": 0.0002, "epoch": 6.4, "step": 2800}, {"loss": 0.7807, "grad_norm": 1.723505973815918, "learning_rate": 0.0002, "epoch": 6.422857142857143, "step": 2810}, {"loss": 0.8645, "grad_norm": 1.4945589303970337, "learning_rate": 0.0002, "epoch": 6.445714285714286, "step": 2820}, {"loss": 0.809, "grad_norm": 1.666458010673523, "learning_rate": 0.0002, "epoch": 6.468571428571429, "step": 2830}, {"loss": 0.7996, "grad_norm": 1.6586525440216064, "learning_rate": 0.0002, "epoch": 6.491428571428571, "step": 2840}, {"loss": 0.8062, "grad_norm": 1.7480043172836304, "learning_rate": 0.0002, "epoch": 6.514285714285714, "step": 2850}, {"loss": 0.7602, "grad_norm": 1.4605649709701538, "learning_rate": 0.0002, "epoch": 6.537142857142857, "step": 2860}, {"loss": 0.8186, "grad_norm": 1.4841814041137695, "learning_rate": 0.0002, "epoch": 6.5600000000000005, "step": 2870}, {"loss": 0.8156, "grad_norm": 1.4653114080429077, "learning_rate": 0.0002, "epoch": 6.582857142857143, "step": 2880}, {"loss": 0.8111, "grad_norm": 1.7266837358474731, "learning_rate": 0.0002, "epoch": 6.605714285714286, "step": 2890}, {"loss": 0.7644, "grad_norm": 1.4860098361968994, "learning_rate": 0.0002, "epoch": 6.628571428571428, "step": 2900}, {"loss": 0.7991, "grad_norm": 1.7177597284317017, "learning_rate": 0.0002, "epoch": 6.651428571428571, "step": 2910}, {"loss": 0.7883, "grad_norm": 1.6757104396820068, "learning_rate": 0.0002, "epoch": 6.674285714285714, "step": 2920}, {"loss": 0.8598, "grad_norm": 1.5177433490753174, "learning_rate": 0.0002, "epoch": 6.6971428571428575, "step": 2930}, {"loss": 0.7825, "grad_norm": 1.8073889017105103, "learning_rate": 0.0002, "epoch": 6.72, "step": 2940}, {"loss": 0.8234, "grad_norm": 1.72337007522583, "learning_rate": 0.0002, "epoch": 6.742857142857143, "step": 2950}, {"loss": 0.896, "grad_norm": 1.6298240423202515, "learning_rate": 0.0002, "epoch": 6.765714285714286, "step": 2960}, {"loss": 0.8252, "grad_norm": 1.6140344142913818, "learning_rate": 0.0002, "epoch": 6.788571428571428, "step": 2970}, {"loss": 0.8314, "grad_norm": 1.7180862426757812, "learning_rate": 0.0002, "epoch": 6.811428571428571, "step": 2980}, {"loss": 0.7929, "grad_norm": 1.7589894533157349, "learning_rate": 0.0002, "epoch": 6.8342857142857145, "step": 2990}, {"loss": 0.828, "grad_norm": 1.780195713043213, "learning_rate": 0.0002, "epoch": 6.857142857142857, "step": 3000}, {"loss": 0.8943, "grad_norm": 1.7182508707046509, "learning_rate": 0.0002, "epoch": 6.88, "step": 3010}, {"loss": 0.7964, "grad_norm": 1.6308406591415405, "learning_rate": 0.0002, "epoch": 6.902857142857143, "step": 3020}, {"loss": 0.8207, "grad_norm": 1.5080229043960571, "learning_rate": 0.0002, "epoch": 6.925714285714285, "step": 3030}, {"loss": 0.886, "grad_norm": 1.623555064201355, "learning_rate": 0.0002, "epoch": 6.948571428571428, "step": 3040}, {"loss": 0.8377, "grad_norm": 1.526054859161377, "learning_rate": 0.0002, "epoch": 6.9714285714285715, "step": 3050}, {"loss": 0.8816, "grad_norm": 1.6671174764633179, "learning_rate": 0.0002, "epoch": 6.994285714285715, "step": 3060}]}
+{"epoch": 0.9988571428571429, "step": 437, "epoch_duration": 1529.840027332306, "total_accumulated_duration": 1529.840027332306, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0606, "grad_norm": 0.7581241726875305, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5819, "grad_norm": 0.528847873210907, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1908, "grad_norm": 0.656889796257019, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1226, "grad_norm": 0.4785534143447876, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.9471, "grad_norm": 0.4260832369327545, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.953, "grad_norm": 0.43787336349487305, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9426, "grad_norm": 0.48819172382354736, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9007, "grad_norm": 0.45916950702667236, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9302, "grad_norm": 0.4116848409175873, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8983, "grad_norm": 1.530875325202942, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9179, "grad_norm": 0.3546597957611084, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8483, "grad_norm": 0.4091620147228241, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7997, "grad_norm": 0.3846702575683594, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.887, "grad_norm": 0.3834623396396637, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.833, "grad_norm": 0.3492090106010437, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.8257, "grad_norm": 0.3689991235733032, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7556, "grad_norm": 0.34498894214630127, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8301, "grad_norm": 0.3145627975463867, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8202, "grad_norm": 0.3054567277431488, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.8217, "grad_norm": 0.3581295907497406, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8552, "grad_norm": 0.35844793915748596, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8673, "grad_norm": 0.3790668547153473, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8689, "grad_norm": 0.36423084139823914, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.853, "grad_norm": 0.32898446917533875, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8766, "grad_norm": 0.35387709736824036, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7915, "grad_norm": 0.31785526871681213, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.8075, "grad_norm": 0.3718429505825043, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.823, "grad_norm": 0.30403199791908264, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8317, "grad_norm": 0.31306174397468567, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.775, "grad_norm": 0.29847824573516846, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8457, "grad_norm": 0.33621150255203247, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8111, "grad_norm": 0.2977754771709442, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7753, "grad_norm": 0.31184402108192444, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8029, "grad_norm": 0.3459164500236511, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7692, "grad_norm": 0.34528955817222595, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.7278, "grad_norm": 0.36549675464630127, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8474, "grad_norm": 0.370948851108551, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.88, "grad_norm": 0.30869928002357483, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8072, "grad_norm": 0.3989962637424469, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7523, "grad_norm": 0.3082427680492401, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7629, "grad_norm": 0.33315128087997437, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8118, "grad_norm": 0.32147616147994995, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7694, "grad_norm": 0.30213138461112976, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}]}
+{"epoch": 7.990857142857143, "step": 3496, "epoch_duration": 1356.2125453948975, "total_accumulated_duration": 10769.926096916199, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-2108-sd-42/checkpoint-875", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.066, "grad_norm": 0.6273946762084961, "learning_rate": 0.0002, "epoch": 0.022857142857142857, "step": 10}, {"loss": 2.5855, "grad_norm": 0.5300710201263428, "learning_rate": 0.0002, "epoch": 0.045714285714285714, "step": 20}, {"loss": 2.1887, "grad_norm": 0.6162196397781372, "learning_rate": 0.0002, "epoch": 0.06857142857142857, "step": 30}, {"loss": 2.1164, "grad_norm": 0.5143047571182251, "learning_rate": 0.0002, "epoch": 0.09142857142857143, "step": 40}, {"loss": 1.943, "grad_norm": 0.4000673294067383, "learning_rate": 0.0002, "epoch": 0.11428571428571428, "step": 50}, {"loss": 1.9531, "grad_norm": 0.444892555475235, "learning_rate": 0.0002, "epoch": 0.13714285714285715, "step": 60}, {"loss": 1.9435, "grad_norm": 0.4871707558631897, "learning_rate": 0.0002, "epoch": 0.16, "step": 70}, {"loss": 1.9072, "grad_norm": 0.451060026884079, "learning_rate": 0.0002, "epoch": 0.18285714285714286, "step": 80}, {"loss": 1.9312, "grad_norm": 0.3939569592475891, "learning_rate": 0.0002, "epoch": 0.2057142857142857, "step": 90}, {"loss": 1.8982, "grad_norm": 0.5033721923828125, "learning_rate": 0.0002, "epoch": 0.22857142857142856, "step": 100}, {"loss": 1.9148, "grad_norm": 0.3636534512042999, "learning_rate": 0.0002, "epoch": 0.25142857142857145, "step": 110}, {"loss": 1.8462, "grad_norm": 0.4391206204891205, "learning_rate": 0.0002, "epoch": 0.2742857142857143, "step": 120}, {"loss": 1.7998, "grad_norm": 0.5243169665336609, "learning_rate": 0.0002, "epoch": 0.29714285714285715, "step": 130}, {"loss": 1.8875, "grad_norm": 0.4055655598640442, "learning_rate": 0.0002, "epoch": 0.32, "step": 140}, {"loss": 1.8348, "grad_norm": 0.39735132455825806, "learning_rate": 0.0002, "epoch": 0.34285714285714286, "step": 150}, {"loss": 1.824, "grad_norm": 0.4696349501609802, "learning_rate": 0.0002, "epoch": 0.3657142857142857, "step": 160}, {"loss": 1.7566, "grad_norm": 0.3987901508808136, "learning_rate": 0.0002, "epoch": 0.38857142857142857, "step": 170}, {"loss": 1.8288, "grad_norm": 0.32404327392578125, "learning_rate": 0.0002, "epoch": 0.4114285714285714, "step": 180}, {"loss": 1.8178, "grad_norm": 0.3692261576652527, "learning_rate": 0.0002, "epoch": 0.4342857142857143, "step": 190}, {"loss": 1.823, "grad_norm": 0.37267744541168213, "learning_rate": 0.0002, "epoch": 0.45714285714285713, "step": 200}, {"loss": 1.8564, "grad_norm": 0.3559934198856354, "learning_rate": 0.0002, "epoch": 0.48, "step": 210}, {"loss": 1.8651, "grad_norm": 0.3374815285205841, "learning_rate": 0.0002, "epoch": 0.5028571428571429, "step": 220}, {"loss": 1.8683, "grad_norm": 0.34598177671432495, "learning_rate": 0.0002, "epoch": 0.5257142857142857, "step": 230}, {"loss": 1.8554, "grad_norm": 0.35629919171333313, "learning_rate": 0.0002, "epoch": 0.5485714285714286, "step": 240}, {"loss": 1.8751, "grad_norm": 0.3586862087249756, "learning_rate": 0.0002, "epoch": 0.5714285714285714, "step": 250}, {"loss": 1.7942, "grad_norm": 0.3198927342891693, "learning_rate": 0.0002, "epoch": 0.5942857142857143, "step": 260}, {"loss": 1.81, "grad_norm": 0.37690025568008423, "learning_rate": 0.0002, "epoch": 0.6171428571428571, "step": 270}, {"loss": 1.8258, "grad_norm": 0.2855667471885681, "learning_rate": 0.0002, "epoch": 0.64, "step": 280}, {"loss": 1.8288, "grad_norm": 0.3242695927619934, "learning_rate": 0.0002, "epoch": 0.6628571428571428, "step": 290}, {"loss": 1.7738, "grad_norm": 0.2960120141506195, "learning_rate": 0.0002, "epoch": 0.6857142857142857, "step": 300}, {"loss": 1.8443, "grad_norm": 0.3596384823322296, "learning_rate": 0.0002, "epoch": 0.7085714285714285, "step": 310}, {"loss": 1.8132, "grad_norm": 0.3001834750175476, "learning_rate": 0.0002, "epoch": 0.7314285714285714, "step": 320}, {"loss": 1.7751, "grad_norm": 0.31361159682273865, "learning_rate": 0.0002, "epoch": 0.7542857142857143, "step": 330}, {"loss": 1.8021, "grad_norm": 0.34093308448791504, "learning_rate": 0.0002, "epoch": 0.7771428571428571, "step": 340}, {"loss": 1.7674, "grad_norm": 0.3383876085281372, "learning_rate": 0.0002, "epoch": 0.8, "step": 350}, {"loss": 1.729, "grad_norm": 0.35100996494293213, "learning_rate": 0.0002, "epoch": 0.8228571428571428, "step": 360}, {"loss": 1.8517, "grad_norm": 0.344976007938385, "learning_rate": 0.0002, "epoch": 0.8457142857142858, "step": 370}, {"loss": 1.8779, "grad_norm": 0.3119729459285736, "learning_rate": 0.0002, "epoch": 0.8685714285714285, "step": 380}, {"loss": 1.8043, "grad_norm": 0.349221795797348, "learning_rate": 0.0002, "epoch": 0.8914285714285715, "step": 390}, {"loss": 1.7529, "grad_norm": 0.3124293386936188, "learning_rate": 0.0002, "epoch": 0.9142857142857143, "step": 400}, {"loss": 1.7611, "grad_norm": 0.35504350066185, "learning_rate": 0.0002, "epoch": 0.9371428571428572, "step": 410}, {"loss": 1.8115, "grad_norm": 0.310310959815979, "learning_rate": 0.0002, "epoch": 0.96, "step": 420}, {"loss": 1.7666, "grad_norm": 0.30432847142219543, "learning_rate": 0.0002, "epoch": 0.9828571428571429, "step": 430}, {"eval_loss": 1.8310153484344482, "eval_runtime": 111.7814, "eval_samples_per_second": 4.536, "eval_steps_per_second": 0.573, "epoch": 0.9988571428571429, "step": 437}, {"loss": 1.7435, "grad_norm": 0.3121616840362549, "learning_rate": 0.0002, "epoch": 1.0057142857142858, "step": 440}, {"loss": 1.7404, "grad_norm": 0.3365118205547333, "learning_rate": 0.0002, "epoch": 1.0285714285714285, "step": 450}, {"loss": 1.7901, "grad_norm": 0.3626686930656433, "learning_rate": 0.0002, "epoch": 1.0514285714285714, "step": 460}, {"loss": 1.721, "grad_norm": 0.30539533495903015, "learning_rate": 0.0002, "epoch": 1.0742857142857143, "step": 470}, {"loss": 1.7354, "grad_norm": 0.3159816861152649, "learning_rate": 0.0002, "epoch": 1.0971428571428572, "step": 480}, {"loss": 1.7471, "grad_norm": 0.3695855736732483, "learning_rate": 0.0002, "epoch": 1.12, "step": 490}, {"loss": 1.7626, "grad_norm": 0.3609161674976349, "learning_rate": 0.0002, "epoch": 1.1428571428571428, "step": 500}, {"loss": 1.8723, "grad_norm": 0.3683869242668152, "learning_rate": 0.0002, "epoch": 1.1657142857142857, "step": 510}, {"loss": 1.7102, "grad_norm": 0.3862539529800415, "learning_rate": 0.0002, "epoch": 1.1885714285714286, "step": 520}, {"loss": 1.7589, "grad_norm": 0.4244740307331085, "learning_rate": 0.0002, "epoch": 1.2114285714285715, "step": 530}, {"loss": 1.717, "grad_norm": 0.373703271150589, "learning_rate": 0.0002, "epoch": 1.2342857142857142, "step": 540}, {"loss": 1.795, "grad_norm": 0.35715773701667786, "learning_rate": 0.0002, "epoch": 1.2571428571428571, "step": 550}, {"loss": 1.7578, "grad_norm": 0.3555964231491089, "learning_rate": 0.0002, "epoch": 1.28, "step": 560}, {"loss": 1.7228, "grad_norm": 0.35080263018608093, "learning_rate": 0.0002, "epoch": 1.302857142857143, "step": 570}, {"loss": 1.6808, "grad_norm": 0.3589482307434082, "learning_rate": 0.0002, "epoch": 1.3257142857142856, "step": 580}, {"loss": 1.7369, "grad_norm": 0.3711223900318146, "learning_rate": 0.0002, "epoch": 1.3485714285714285, "step": 590}, {"loss": 1.7417, "grad_norm": 0.313614159822464, "learning_rate": 0.0002, "epoch": 1.3714285714285714, "step": 600}, {"loss": 1.7191, "grad_norm": 0.3842357397079468, "learning_rate": 0.0002, "epoch": 1.3942857142857144, "step": 610}, {"loss": 1.6737, "grad_norm": 0.36126819252967834, "learning_rate": 0.0002, "epoch": 1.4171428571428573, "step": 620}, {"loss": 1.7073, "grad_norm": 0.35922661423683167, "learning_rate": 0.0002, "epoch": 1.44, "step": 630}, {"loss": 1.6708, "grad_norm": 0.3922875225543976, "learning_rate": 0.0002, "epoch": 1.4628571428571429, "step": 640}, {"loss": 1.7544, "grad_norm": 0.365546852350235, "learning_rate": 0.0002, "epoch": 1.4857142857142858, "step": 650}, {"loss": 1.674, "grad_norm": 0.36107590794563293, "learning_rate": 0.0002, "epoch": 1.5085714285714285, "step": 660}, {"loss": 1.6518, "grad_norm": 0.3307042121887207, "learning_rate": 0.0002, "epoch": 1.5314285714285716, "step": 670}, {"loss": 1.7306, "grad_norm": 0.3492133915424347, "learning_rate": 0.0002, "epoch": 1.5542857142857143, "step": 680}, {"loss": 1.812, "grad_norm": 0.38608574867248535, "learning_rate": 0.0002, "epoch": 1.5771428571428572, "step": 690}, {"loss": 1.7042, "grad_norm": 0.3489173650741577, "learning_rate": 0.0002, "epoch": 1.6, "step": 700}, {"loss": 1.7306, "grad_norm": 0.36614152789115906, "learning_rate": 0.0002, "epoch": 1.6228571428571428, "step": 710}, {"loss": 1.7281, "grad_norm": 0.34340205788612366, "learning_rate": 0.0002, "epoch": 1.6457142857142857, "step": 720}, {"loss": 1.7607, "grad_norm": 0.34590771794319153, "learning_rate": 0.0002, "epoch": 1.6685714285714286, "step": 730}, {"loss": 1.7108, "grad_norm": 0.3759954273700714, "learning_rate": 0.0002, "epoch": 1.6914285714285713, "step": 740}, {"loss": 1.6903, "grad_norm": 0.3753475546836853, "learning_rate": 0.0002, "epoch": 1.7142857142857144, "step": 750}, {"loss": 1.7054, "grad_norm": 0.38416001200675964, "learning_rate": 0.0002, "epoch": 1.737142857142857, "step": 760}, {"loss": 1.7125, "grad_norm": 0.36223554611206055, "learning_rate": 0.0002, "epoch": 1.76, "step": 770}, {"loss": 1.6878, "grad_norm": 0.329556941986084, "learning_rate": 0.0002, "epoch": 1.782857142857143, "step": 780}, {"loss": 1.803, "grad_norm": 0.34008052945137024, "learning_rate": 0.0002, "epoch": 1.8057142857142856, "step": 790}, {"loss": 1.6545, "grad_norm": 0.40297919511795044, "learning_rate": 0.0002, "epoch": 1.8285714285714287, "step": 800}, {"loss": 1.7518, "grad_norm": 0.35378390550613403, "learning_rate": 0.0002, "epoch": 1.8514285714285714, "step": 810}, {"loss": 1.7063, "grad_norm": 0.3625478148460388, "learning_rate": 0.0002, "epoch": 1.8742857142857143, "step": 820}, {"loss": 1.6339, "grad_norm": 0.36153221130371094, "learning_rate": 0.0002, "epoch": 1.8971428571428572, "step": 830}, {"loss": 1.7434, "grad_norm": 0.3612948954105377, "learning_rate": 0.0002, "epoch": 1.92, "step": 840}, {"loss": 1.7242, "grad_norm": 0.399213045835495, "learning_rate": 0.0002, "epoch": 1.9428571428571428, "step": 850}, {"loss": 1.6909, "grad_norm": 0.40026402473449707, "learning_rate": 0.0002, "epoch": 1.9657142857142857, "step": 860}, {"loss": 1.76, "grad_norm": 0.38114118576049805, "learning_rate": 0.0002, "epoch": 1.9885714285714284, "step": 870}, {"eval_loss": 1.8279441595077515, "eval_runtime": 111.1652, "eval_samples_per_second": 4.561, "eval_steps_per_second": 0.576, "epoch": 2.0, "step": 875}, {"loss": 1.7215, "grad_norm": 0.33838793635368347, "learning_rate": 0.0002, "epoch": 2.0114285714285716, "step": 880}, {"loss": 1.6105, "grad_norm": 0.4219334125518799, "learning_rate": 0.0002, "epoch": 2.0342857142857143, "step": 890}, {"loss": 1.6144, "grad_norm": 0.43962377309799194, "learning_rate": 0.0002, "epoch": 2.057142857142857, "step": 900}, {"loss": 1.5821, "grad_norm": 0.41956576704978943, "learning_rate": 0.0002, "epoch": 2.08, "step": 910}, {"loss": 1.5812, "grad_norm": 0.4439629912376404, "learning_rate": 0.0002, "epoch": 2.1028571428571428, "step": 920}, {"loss": 1.5923, "grad_norm": 0.43405696749687195, "learning_rate": 0.0002, "epoch": 2.125714285714286, "step": 930}, {"loss": 1.5968, "grad_norm": 0.4321737587451935, "learning_rate": 0.0002, "epoch": 2.1485714285714286, "step": 940}, {"loss": 1.5683, "grad_norm": 0.4689100682735443, "learning_rate": 0.0002, "epoch": 2.1714285714285713, "step": 950}, {"loss": 1.6442, "grad_norm": 0.47024697065353394, "learning_rate": 0.0002, "epoch": 2.1942857142857144, "step": 960}, {"loss": 1.4703, "grad_norm": 0.4535103440284729, "learning_rate": 0.0002, "epoch": 2.217142857142857, "step": 970}, {"loss": 1.6378, "grad_norm": 0.45990121364593506, "learning_rate": 0.0002, "epoch": 2.24, "step": 980}, {"loss": 1.627, "grad_norm": 0.48427215218544006, "learning_rate": 0.0002, "epoch": 2.262857142857143, "step": 990}, {"loss": 1.6316, "grad_norm": 0.43076643347740173, "learning_rate": 0.0002, "epoch": 2.2857142857142856, "step": 1000}, {"loss": 1.5938, "grad_norm": 0.4854483902454376, "learning_rate": 0.0002, "epoch": 2.3085714285714287, "step": 1010}, {"loss": 1.6223, "grad_norm": 0.46086496114730835, "learning_rate": 0.0002, "epoch": 2.3314285714285714, "step": 1020}, {"loss": 1.6392, "grad_norm": 0.4714847505092621, "learning_rate": 0.0002, "epoch": 2.354285714285714, "step": 1030}, {"loss": 1.5785, "grad_norm": 0.4423409402370453, "learning_rate": 0.0002, "epoch": 2.3771428571428572, "step": 1040}, {"loss": 1.5821, "grad_norm": 0.46261295676231384, "learning_rate": 0.0002, "epoch": 2.4, "step": 1050}, {"loss": 1.5503, "grad_norm": 0.4914337396621704, "learning_rate": 0.0002, "epoch": 2.422857142857143, "step": 1060}, {"loss": 1.6459, "grad_norm": 0.45144036412239075, "learning_rate": 0.0002, "epoch": 2.4457142857142857, "step": 1070}, {"loss": 1.6416, "grad_norm": 0.4510825276374817, "learning_rate": 0.0002, "epoch": 2.4685714285714284, "step": 1080}, {"loss": 1.5808, "grad_norm": 0.48552489280700684, "learning_rate": 0.0002, "epoch": 2.4914285714285715, "step": 1090}, {"loss": 1.6659, "grad_norm": 0.4768163859844208, "learning_rate": 0.0002, "epoch": 2.5142857142857142, "step": 1100}, {"loss": 1.6251, "grad_norm": 0.5192609429359436, "learning_rate": 0.0002, "epoch": 2.5371428571428574, "step": 1110}, {"loss": 1.5581, "grad_norm": 0.49308598041534424, "learning_rate": 0.0002, "epoch": 2.56, "step": 1120}, {"loss": 1.6767, "grad_norm": 0.5068584084510803, "learning_rate": 0.0002, "epoch": 2.5828571428571427, "step": 1130}, {"loss": 1.5788, "grad_norm": 0.4822661280632019, "learning_rate": 0.0002, "epoch": 2.605714285714286, "step": 1140}, {"loss": 1.6891, "grad_norm": 0.5028144717216492, "learning_rate": 0.0002, "epoch": 2.6285714285714286, "step": 1150}, {"loss": 1.6782, "grad_norm": 0.48315200209617615, "learning_rate": 0.0002, "epoch": 2.6514285714285712, "step": 1160}, {"loss": 1.5508, "grad_norm": 0.551934540271759, "learning_rate": 0.0002, "epoch": 2.6742857142857144, "step": 1170}, {"loss": 1.5448, "grad_norm": 0.49223729968070984, "learning_rate": 0.0002, "epoch": 2.697142857142857, "step": 1180}, {"loss": 1.6382, "grad_norm": 0.514847457408905, "learning_rate": 0.0002, "epoch": 2.7199999999999998, "step": 1190}, {"loss": 1.5054, "grad_norm": 0.4830605387687683, "learning_rate": 0.0002, "epoch": 2.742857142857143, "step": 1200}, {"loss": 1.6285, "grad_norm": 0.4584822952747345, "learning_rate": 0.0002, "epoch": 2.7657142857142856, "step": 1210}, {"loss": 1.6043, "grad_norm": 0.4688762426376343, "learning_rate": 0.0002, "epoch": 2.7885714285714287, "step": 1220}, {"loss": 1.5555, "grad_norm": 0.4488156735897064, "learning_rate": 0.0002, "epoch": 2.8114285714285714, "step": 1230}, {"loss": 1.5523, "grad_norm": 0.4700278639793396, "learning_rate": 0.0002, "epoch": 2.8342857142857145, "step": 1240}, {"loss": 1.642, "grad_norm": 0.5282207131385803, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 1250}, {"loss": 1.5532, "grad_norm": 0.4874219000339508, "learning_rate": 0.0002, "epoch": 2.88, "step": 1260}, {"loss": 1.6051, "grad_norm": 0.49468332529067993, "learning_rate": 0.0002, "epoch": 2.902857142857143, "step": 1270}, {"loss": 1.6006, "grad_norm": 0.49770233035087585, "learning_rate": 0.0002, "epoch": 2.9257142857142857, "step": 1280}, {"loss": 1.606, "grad_norm": 0.4433252811431885, "learning_rate": 0.0002, "epoch": 2.9485714285714284, "step": 1290}, {"loss": 1.608, "grad_norm": 0.46836379170417786, "learning_rate": 0.0002, "epoch": 2.9714285714285715, "step": 1300}, {"loss": 1.5555, "grad_norm": 0.5001904368400574, "learning_rate": 0.0002, "epoch": 2.994285714285714, "step": 1310}, {"eval_loss": 1.8684407472610474, "eval_runtime": 111.2835, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 2.998857142857143, "step": 1312}, {"loss": 1.4486, "grad_norm": 0.5184893012046814, "learning_rate": 0.0002, "epoch": 3.0171428571428573, "step": 1320}, {"loss": 1.4082, "grad_norm": 0.5665355920791626, "learning_rate": 0.0002, "epoch": 3.04, "step": 1330}, {"loss": 1.3741, "grad_norm": 0.6601403951644897, "learning_rate": 0.0002, "epoch": 3.0628571428571427, "step": 1340}, {"loss": 1.433, "grad_norm": 0.6921621561050415, "learning_rate": 0.0002, "epoch": 3.085714285714286, "step": 1350}, {"loss": 1.4562, "grad_norm": 0.6406348943710327, "learning_rate": 0.0002, "epoch": 3.1085714285714285, "step": 1360}, {"loss": 1.3563, "grad_norm": 0.5814554691314697, "learning_rate": 0.0002, "epoch": 3.1314285714285712, "step": 1370}, {"loss": 1.4096, "grad_norm": 0.683325469493866, "learning_rate": 0.0002, "epoch": 3.1542857142857144, "step": 1380}, {"loss": 1.4106, "grad_norm": 0.6686155200004578, "learning_rate": 0.0002, "epoch": 3.177142857142857, "step": 1390}, {"loss": 1.4394, "grad_norm": 0.8159713745117188, "learning_rate": 0.0002, "epoch": 3.2, "step": 1400}, {"loss": 1.4279, "grad_norm": 0.646216094493866, "learning_rate": 0.0002, "epoch": 3.222857142857143, "step": 1410}, {"loss": 1.4232, "grad_norm": 0.7323529720306396, "learning_rate": 0.0002, "epoch": 3.2457142857142856, "step": 1420}, {"loss": 1.3891, "grad_norm": 0.689349353313446, "learning_rate": 0.0002, "epoch": 3.2685714285714287, "step": 1430}, {"loss": 1.4578, "grad_norm": 0.727894127368927, "learning_rate": 0.0002, "epoch": 3.2914285714285714, "step": 1440}, {"loss": 1.4, "grad_norm": 0.6921590566635132, "learning_rate": 0.0002, "epoch": 3.314285714285714, "step": 1450}, {"loss": 1.4272, "grad_norm": 0.6176243424415588, "learning_rate": 0.0002, "epoch": 3.337142857142857, "step": 1460}, {"loss": 1.4323, "grad_norm": 0.9006354212760925, "learning_rate": 0.0002, "epoch": 3.36, "step": 1470}, {"loss": 1.4353, "grad_norm": 0.8145929574966431, "learning_rate": 0.0002, "epoch": 3.382857142857143, "step": 1480}, {"loss": 1.3859, "grad_norm": 0.6640016436576843, "learning_rate": 0.0002, "epoch": 3.4057142857142857, "step": 1490}, {"loss": 1.387, "grad_norm": 0.7266780138015747, "learning_rate": 0.0002, "epoch": 3.4285714285714284, "step": 1500}, {"loss": 1.4108, "grad_norm": 0.9351356029510498, "learning_rate": 0.0002, "epoch": 3.4514285714285715, "step": 1510}, {"loss": 1.4656, "grad_norm": 0.675645649433136, "learning_rate": 0.0002, "epoch": 3.474285714285714, "step": 1520}, {"loss": 1.384, "grad_norm": 0.761472225189209, "learning_rate": 0.0002, "epoch": 3.4971428571428573, "step": 1530}, {"loss": 1.4968, "grad_norm": 0.6653069257736206, "learning_rate": 0.0002, "epoch": 3.52, "step": 1540}, {"loss": 1.4686, "grad_norm": 0.667412519454956, "learning_rate": 0.0002, "epoch": 3.5428571428571427, "step": 1550}, {"loss": 1.4241, "grad_norm": 0.6395593881607056, "learning_rate": 0.0002, "epoch": 3.565714285714286, "step": 1560}, {"loss": 1.4825, "grad_norm": 0.7588621377944946, "learning_rate": 0.0002, "epoch": 3.5885714285714285, "step": 1570}, {"loss": 1.4459, "grad_norm": 0.6206456422805786, "learning_rate": 0.0002, "epoch": 3.611428571428571, "step": 1580}, {"loss": 1.436, "grad_norm": 0.7591291666030884, "learning_rate": 0.0002, "epoch": 3.6342857142857143, "step": 1590}, {"loss": 1.458, "grad_norm": 0.6476313471794128, "learning_rate": 0.0002, "epoch": 3.657142857142857, "step": 1600}, {"loss": 1.4598, "grad_norm": 0.6731392741203308, "learning_rate": 0.0002, "epoch": 3.68, "step": 1610}, {"loss": 1.4225, "grad_norm": 0.725190281867981, "learning_rate": 0.0002, "epoch": 3.702857142857143, "step": 1620}, {"loss": 1.4525, "grad_norm": 0.6720049977302551, "learning_rate": 0.0002, "epoch": 3.725714285714286, "step": 1630}, {"loss": 1.429, "grad_norm": 0.6301007270812988, "learning_rate": 0.0002, "epoch": 3.7485714285714287, "step": 1640}, {"loss": 1.4166, "grad_norm": 0.715893566608429, "learning_rate": 0.0002, "epoch": 3.7714285714285714, "step": 1650}, {"loss": 1.3624, "grad_norm": 0.7539359927177429, "learning_rate": 0.0002, "epoch": 3.7942857142857145, "step": 1660}, {"loss": 1.4516, "grad_norm": 0.6658543348312378, "learning_rate": 0.0002, "epoch": 3.817142857142857, "step": 1670}, {"loss": 1.3934, "grad_norm": 0.7019526958465576, "learning_rate": 0.0002, "epoch": 3.84, "step": 1680}, {"loss": 1.4436, "grad_norm": 0.6517802476882935, "learning_rate": 0.0002, "epoch": 3.862857142857143, "step": 1690}, {"loss": 1.4968, "grad_norm": 0.7617332935333252, "learning_rate": 0.0002, "epoch": 3.8857142857142857, "step": 1700}, {"loss": 1.5145, "grad_norm": 0.6919480562210083, "learning_rate": 0.0002, "epoch": 3.9085714285714284, "step": 1710}, {"loss": 1.4317, "grad_norm": 0.6987943053245544, "learning_rate": 0.0002, "epoch": 3.9314285714285715, "step": 1720}, {"loss": 1.4704, "grad_norm": 0.7062228918075562, "learning_rate": 0.0002, "epoch": 3.954285714285714, "step": 1730}, {"loss": 1.4219, "grad_norm": 0.6769542098045349, "learning_rate": 0.0002, "epoch": 3.977142857142857, "step": 1740}, {"loss": 1.4998, "grad_norm": 0.6832144260406494, "learning_rate": 0.0002, "epoch": 4.0, "step": 1750}, {"eval_loss": 1.9474865198135376, "eval_runtime": 111.288, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.0, "step": 1750}, {"loss": 1.2251, "grad_norm": 1.064110279083252, "learning_rate": 0.0002, "epoch": 4.022857142857143, "step": 1760}, {"loss": 1.2013, "grad_norm": 0.8380683660507202, "learning_rate": 0.0002, "epoch": 4.045714285714285, "step": 1770}, {"loss": 1.2416, "grad_norm": 1.1863020658493042, "learning_rate": 0.0002, "epoch": 4.0685714285714285, "step": 1780}, {"loss": 1.2499, "grad_norm": 1.0128898620605469, "learning_rate": 0.0002, "epoch": 4.091428571428572, "step": 1790}, {"loss": 1.2043, "grad_norm": 0.9221312403678894, "learning_rate": 0.0002, "epoch": 4.114285714285714, "step": 1800}, {"loss": 1.181, "grad_norm": 1.1298727989196777, "learning_rate": 0.0002, "epoch": 4.137142857142857, "step": 1810}, {"loss": 1.1491, "grad_norm": 0.8854547739028931, "learning_rate": 0.0002, "epoch": 4.16, "step": 1820}, {"loss": 1.2156, "grad_norm": 0.8920808434486389, "learning_rate": 0.0002, "epoch": 4.182857142857143, "step": 1830}, {"loss": 1.1969, "grad_norm": 0.913244366645813, "learning_rate": 0.0002, "epoch": 4.2057142857142855, "step": 1840}, {"loss": 1.2156, "grad_norm": 0.908831000328064, "learning_rate": 0.0002, "epoch": 4.228571428571429, "step": 1850}, {"loss": 1.1653, "grad_norm": 1.0223685503005981, "learning_rate": 0.0002, "epoch": 4.251428571428572, "step": 1860}, {"loss": 1.2497, "grad_norm": 0.9771921634674072, "learning_rate": 0.0002, "epoch": 4.274285714285714, "step": 1870}, {"loss": 1.213, "grad_norm": 0.9313384890556335, "learning_rate": 0.0002, "epoch": 4.297142857142857, "step": 1880}, {"loss": 1.1723, "grad_norm": 1.0754257440567017, "learning_rate": 0.0002, "epoch": 4.32, "step": 1890}, {"loss": 1.2286, "grad_norm": 0.8904672265052795, "learning_rate": 0.0002, "epoch": 4.3428571428571425, "step": 1900}, {"loss": 1.2618, "grad_norm": 1.046527624130249, "learning_rate": 0.0002, "epoch": 4.365714285714286, "step": 1910}, {"loss": 1.2368, "grad_norm": 0.9576982855796814, "learning_rate": 0.0002, "epoch": 4.388571428571429, "step": 1920}, {"loss": 1.211, "grad_norm": 0.9278356432914734, "learning_rate": 0.0002, "epoch": 4.411428571428571, "step": 1930}, {"loss": 1.2005, "grad_norm": 1.1763030290603638, "learning_rate": 0.0002, "epoch": 4.434285714285714, "step": 1940}, {"loss": 1.1541, "grad_norm": 0.9183000326156616, "learning_rate": 0.0002, "epoch": 4.457142857142857, "step": 1950}, {"loss": 1.2257, "grad_norm": 1.050980806350708, "learning_rate": 0.0002, "epoch": 4.48, "step": 1960}, {"loss": 1.2133, "grad_norm": 0.9975392818450928, "learning_rate": 0.0002, "epoch": 4.502857142857143, "step": 1970}, {"loss": 1.2312, "grad_norm": 0.990544319152832, "learning_rate": 0.0002, "epoch": 4.525714285714286, "step": 1980}, {"loss": 1.2465, "grad_norm": 1.004794955253601, "learning_rate": 0.0002, "epoch": 4.548571428571429, "step": 1990}, {"loss": 1.2085, "grad_norm": 0.9294857978820801, "learning_rate": 0.0002, "epoch": 4.571428571428571, "step": 2000}, {"loss": 1.2874, "grad_norm": 0.93436598777771, "learning_rate": 0.0002, "epoch": 4.594285714285714, "step": 2010}, {"loss": 1.1965, "grad_norm": 0.8704655766487122, "learning_rate": 0.0002, "epoch": 4.617142857142857, "step": 2020}, {"loss": 1.204, "grad_norm": 0.9077927470207214, "learning_rate": 0.0002, "epoch": 4.64, "step": 2030}, {"loss": 1.2198, "grad_norm": 0.912987470626831, "learning_rate": 0.0002, "epoch": 4.662857142857143, "step": 2040}, {"loss": 1.2868, "grad_norm": 0.9740643501281738, "learning_rate": 0.0002, "epoch": 4.685714285714286, "step": 2050}, {"loss": 1.249, "grad_norm": 1.133357048034668, "learning_rate": 0.0002, "epoch": 4.708571428571428, "step": 2060}, {"loss": 1.1974, "grad_norm": 0.8844527006149292, "learning_rate": 0.0002, "epoch": 4.731428571428571, "step": 2070}, {"loss": 1.2481, "grad_norm": 1.0083311796188354, "learning_rate": 0.0002, "epoch": 4.7542857142857144, "step": 2080}, {"loss": 1.263, "grad_norm": 1.000447154045105, "learning_rate": 0.0002, "epoch": 4.777142857142858, "step": 2090}, {"loss": 1.2313, "grad_norm": 0.9620300531387329, "learning_rate": 0.0002, "epoch": 4.8, "step": 2100}, {"loss": 1.2659, "grad_norm": 0.9843335151672363, "learning_rate": 0.0002, "epoch": 4.822857142857143, "step": 2110}, {"loss": 1.2535, "grad_norm": 0.9906681180000305, "learning_rate": 0.0002, "epoch": 4.845714285714286, "step": 2120}, {"loss": 1.2325, "grad_norm": 0.9544073939323425, "learning_rate": 0.0002, "epoch": 4.868571428571428, "step": 2130}, {"loss": 1.284, "grad_norm": 0.9392994046211243, "learning_rate": 0.0002, "epoch": 4.8914285714285715, "step": 2140}, {"loss": 1.3075, "grad_norm": 1.104519248008728, "learning_rate": 0.0002, "epoch": 4.914285714285715, "step": 2150}, {"loss": 1.2753, "grad_norm": 0.9495956897735596, "learning_rate": 0.0002, "epoch": 4.937142857142857, "step": 2160}, {"loss": 1.2412, "grad_norm": 0.9696287512779236, "learning_rate": 0.0002, "epoch": 4.96, "step": 2170}, {"loss": 1.2354, "grad_norm": 0.9933681488037109, "learning_rate": 0.0002, "epoch": 4.982857142857143, "step": 2180}, {"eval_loss": 2.099808692932129, "eval_runtime": 111.2808, "eval_samples_per_second": 4.556, "eval_steps_per_second": 0.575, "epoch": 4.998857142857143, "step": 2187}, {"loss": 1.2183, "grad_norm": 0.9482853412628174, "learning_rate": 0.0002, "epoch": 5.005714285714285, "step": 2190}, {"loss": 0.9898, "grad_norm": 1.6689555644989014, "learning_rate": 0.0002, "epoch": 5.0285714285714285, "step": 2200}, {"loss": 0.9741, "grad_norm": 1.2019699811935425, "learning_rate": 0.0002, "epoch": 5.051428571428572, "step": 2210}, {"loss": 0.9737, "grad_norm": 1.535780429840088, "learning_rate": 0.0002, "epoch": 5.074285714285715, "step": 2220}, {"loss": 0.9494, "grad_norm": 1.2061309814453125, "learning_rate": 0.0002, "epoch": 5.097142857142857, "step": 2230}, {"loss": 0.9316, "grad_norm": 1.1898778676986694, "learning_rate": 0.0002, "epoch": 5.12, "step": 2240}, {"loss": 1.002, "grad_norm": 1.158898949623108, "learning_rate": 0.0002, "epoch": 5.142857142857143, "step": 2250}, {"loss": 0.9715, "grad_norm": 1.370749592781067, "learning_rate": 0.0002, "epoch": 5.1657142857142855, "step": 2260}, {"loss": 0.9365, "grad_norm": 1.314120888710022, "learning_rate": 0.0002, "epoch": 5.188571428571429, "step": 2270}, {"loss": 1.0316, "grad_norm": 1.2184966802597046, "learning_rate": 0.0002, "epoch": 5.211428571428572, "step": 2280}, {"loss": 0.9407, "grad_norm": 1.4833279848098755, "learning_rate": 0.0002, "epoch": 5.234285714285714, "step": 2290}, {"loss": 0.9635, "grad_norm": 1.3348219394683838, "learning_rate": 0.0002, "epoch": 5.257142857142857, "step": 2300}, {"loss": 1.0294, "grad_norm": 1.4166619777679443, "learning_rate": 0.0002, "epoch": 5.28, "step": 2310}, {"loss": 0.9818, "grad_norm": 1.4539530277252197, "learning_rate": 0.0002, "epoch": 5.3028571428571425, "step": 2320}, {"loss": 1.0165, "grad_norm": 1.4642518758773804, "learning_rate": 0.0002, "epoch": 5.325714285714286, "step": 2330}, {"loss": 1.0081, "grad_norm": 1.3938848972320557, "learning_rate": 0.0002, "epoch": 5.348571428571429, "step": 2340}, {"loss": 1.03, "grad_norm": 1.1147894859313965, "learning_rate": 0.0002, "epoch": 5.371428571428572, "step": 2350}, {"loss": 0.9975, "grad_norm": 1.3465309143066406, "learning_rate": 0.0002, "epoch": 5.394285714285714, "step": 2360}, {"loss": 1.0138, "grad_norm": 1.4788566827774048, "learning_rate": 0.0002, "epoch": 5.417142857142857, "step": 2370}, {"loss": 0.9896, "grad_norm": 1.3808705806732178, "learning_rate": 0.0002, "epoch": 5.44, "step": 2380}, {"loss": 1.0279, "grad_norm": 1.2336329221725464, "learning_rate": 0.0002, "epoch": 5.462857142857143, "step": 2390}, {"loss": 0.9763, "grad_norm": 1.5445678234100342, "learning_rate": 0.0002, "epoch": 5.485714285714286, "step": 2400}, {"loss": 0.9534, "grad_norm": 1.107488989830017, "learning_rate": 0.0002, "epoch": 5.508571428571429, "step": 2410}, {"loss": 1.0036, "grad_norm": 1.39687979221344, "learning_rate": 0.0002, "epoch": 5.531428571428571, "step": 2420}, {"loss": 0.9959, "grad_norm": 1.3905695676803589, "learning_rate": 0.0002, "epoch": 5.554285714285714, "step": 2430}, {"loss": 0.9912, "grad_norm": 1.3772821426391602, "learning_rate": 0.0002, "epoch": 5.577142857142857, "step": 2440}, {"loss": 0.9825, "grad_norm": 1.1661899089813232, "learning_rate": 0.0002, "epoch": 5.6, "step": 2450}, {"loss": 1.0003, "grad_norm": 1.2730463743209839, "learning_rate": 0.0002, "epoch": 5.622857142857143, "step": 2460}, {"loss": 1.0433, "grad_norm": 1.2251193523406982, "learning_rate": 0.0002, "epoch": 5.645714285714286, "step": 2470}, {"loss": 1.079, "grad_norm": 1.5454859733581543, "learning_rate": 0.0002, "epoch": 5.668571428571429, "step": 2480}, {"loss": 1.0414, "grad_norm": 1.5405735969543457, "learning_rate": 0.0002, "epoch": 5.691428571428571, "step": 2490}, {"loss": 1.0353, "grad_norm": 1.2555434703826904, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 2500}, {"loss": 1.0019, "grad_norm": 1.3323487043380737, "learning_rate": 0.0002, "epoch": 5.737142857142857, "step": 2510}, {"loss": 1.051, "grad_norm": 1.3106356859207153, "learning_rate": 0.0002, "epoch": 5.76, "step": 2520}, {"loss": 1.0248, "grad_norm": 1.4832439422607422, "learning_rate": 0.0002, "epoch": 5.782857142857143, "step": 2530}, {"loss": 1.0643, "grad_norm": 1.1336562633514404, "learning_rate": 0.0002, "epoch": 5.805714285714286, "step": 2540}, {"loss": 1.0446, "grad_norm": 1.2434223890304565, "learning_rate": 0.0002, "epoch": 5.828571428571428, "step": 2550}, {"loss": 1.0467, "grad_norm": 1.2825450897216797, "learning_rate": 0.0002, "epoch": 5.851428571428571, "step": 2560}, {"loss": 1.0642, "grad_norm": 1.4373180866241455, "learning_rate": 0.0002, "epoch": 5.8742857142857146, "step": 2570}, {"loss": 1.0814, "grad_norm": 1.435015320777893, "learning_rate": 0.0002, "epoch": 5.897142857142857, "step": 2580}, {"loss": 1.0272, "grad_norm": 1.4075653553009033, "learning_rate": 0.0002, "epoch": 5.92, "step": 2590}, {"loss": 1.0703, "grad_norm": 1.319630742073059, "learning_rate": 0.0002, "epoch": 5.942857142857143, "step": 2600}, {"loss": 1.0375, "grad_norm": 1.278330683708191, "learning_rate": 0.0002, "epoch": 5.965714285714286, "step": 2610}, {"loss": 1.0766, "grad_norm": 1.258158564567566, "learning_rate": 0.0002, "epoch": 5.988571428571428, "step": 2620}, {"eval_loss": 2.3689301013946533, "eval_runtime": 53.9067, "eval_samples_per_second": 9.405, "eval_steps_per_second": 1.187, "epoch": 6.0, "step": 2625}, {"loss": 0.9142, "grad_norm": 1.3128368854522705, "learning_rate": 0.0002, "epoch": 6.011428571428572, "step": 2630}, {"loss": 0.7716, "grad_norm": 1.4280474185943604, "learning_rate": 0.0002, "epoch": 6.034285714285715, "step": 2640}, {"loss": 0.7776, "grad_norm": 1.5061450004577637, "learning_rate": 0.0002, "epoch": 6.057142857142857, "step": 2650}, {"loss": 0.7707, "grad_norm": 1.6013342142105103, "learning_rate": 0.0002, "epoch": 6.08, "step": 2660}, {"loss": 0.7543, "grad_norm": 2.0107381343841553, "learning_rate": 0.0002, "epoch": 6.102857142857143, "step": 2670}, {"loss": 0.747, "grad_norm": 1.5010124444961548, "learning_rate": 0.0002, "epoch": 6.1257142857142854, "step": 2680}, {"loss": 0.7501, "grad_norm": 1.5222150087356567, "learning_rate": 0.0002, "epoch": 6.148571428571429, "step": 2690}, {"loss": 0.7712, "grad_norm": 1.5413103103637695, "learning_rate": 0.0002, "epoch": 6.171428571428572, "step": 2700}, {"loss": 0.7, "grad_norm": 1.527140736579895, "learning_rate": 0.0002, "epoch": 6.194285714285714, "step": 2710}, {"loss": 0.7539, "grad_norm": 1.9386590719223022, "learning_rate": 0.0002, "epoch": 6.217142857142857, "step": 2720}, {"loss": 0.7586, "grad_norm": 1.8115214109420776, "learning_rate": 0.0002, "epoch": 6.24, "step": 2730}, {"loss": 0.7426, "grad_norm": 1.6221802234649658, "learning_rate": 0.0002, "epoch": 6.2628571428571425, "step": 2740}, {"loss": 0.8002, "grad_norm": 1.6698768138885498, "learning_rate": 0.0002, "epoch": 6.285714285714286, "step": 2750}, {"loss": 0.7293, "grad_norm": 1.7960610389709473, "learning_rate": 0.0002, "epoch": 6.308571428571429, "step": 2760}, {"loss": 0.7405, "grad_norm": 1.32172429561615, "learning_rate": 0.0002, "epoch": 6.331428571428571, "step": 2770}, {"loss": 0.7198, "grad_norm": 1.7468090057373047, "learning_rate": 0.0002, "epoch": 6.354285714285714, "step": 2780}, {"loss": 0.76, "grad_norm": 1.6777397394180298, "learning_rate": 0.0002, "epoch": 6.377142857142857, "step": 2790}, {"loss": 0.7879, "grad_norm": 1.6200671195983887, "learning_rate": 0.0002, "epoch": 6.4, "step": 2800}, {"loss": 0.7807, "grad_norm": 1.723505973815918, "learning_rate": 0.0002, "epoch": 6.422857142857143, "step": 2810}, {"loss": 0.8645, "grad_norm": 1.4945589303970337, "learning_rate": 0.0002, "epoch": 6.445714285714286, "step": 2820}, {"loss": 0.809, "grad_norm": 1.666458010673523, "learning_rate": 0.0002, "epoch": 6.468571428571429, "step": 2830}, {"loss": 0.7996, "grad_norm": 1.6586525440216064, "learning_rate": 0.0002, "epoch": 6.491428571428571, "step": 2840}, {"loss": 0.8062, "grad_norm": 1.7480043172836304, "learning_rate": 0.0002, "epoch": 6.514285714285714, "step": 2850}, {"loss": 0.7602, "grad_norm": 1.4605649709701538, "learning_rate": 0.0002, "epoch": 6.537142857142857, "step": 2860}, {"loss": 0.8186, "grad_norm": 1.4841814041137695, "learning_rate": 0.0002, "epoch": 6.5600000000000005, "step": 2870}, {"loss": 0.8156, "grad_norm": 1.4653114080429077, "learning_rate": 0.0002, "epoch": 6.582857142857143, "step": 2880}, {"loss": 0.8111, "grad_norm": 1.7266837358474731, "learning_rate": 0.0002, "epoch": 6.605714285714286, "step": 2890}, {"loss": 0.7644, "grad_norm": 1.4860098361968994, "learning_rate": 0.0002, "epoch": 6.628571428571428, "step": 2900}, {"loss": 0.7991, "grad_norm": 1.7177597284317017, "learning_rate": 0.0002, "epoch": 6.651428571428571, "step": 2910}, {"loss": 0.7883, "grad_norm": 1.6757104396820068, "learning_rate": 0.0002, "epoch": 6.674285714285714, "step": 2920}, {"loss": 0.8598, "grad_norm": 1.5177433490753174, "learning_rate": 0.0002, "epoch": 6.6971428571428575, "step": 2930}, {"loss": 0.7825, "grad_norm": 1.8073889017105103, "learning_rate": 0.0002, "epoch": 6.72, "step": 2940}, {"loss": 0.8234, "grad_norm": 1.72337007522583, "learning_rate": 0.0002, "epoch": 6.742857142857143, "step": 2950}, {"loss": 0.896, "grad_norm": 1.6298240423202515, "learning_rate": 0.0002, "epoch": 6.765714285714286, "step": 2960}, {"loss": 0.8252, "grad_norm": 1.6140344142913818, "learning_rate": 0.0002, "epoch": 6.788571428571428, "step": 2970}, {"loss": 0.8314, "grad_norm": 1.7180862426757812, "learning_rate": 0.0002, "epoch": 6.811428571428571, "step": 2980}, {"loss": 0.7929, "grad_norm": 1.7589894533157349, "learning_rate": 0.0002, "epoch": 6.8342857142857145, "step": 2990}, {"loss": 0.828, "grad_norm": 1.780195713043213, "learning_rate": 0.0002, "epoch": 6.857142857142857, "step": 3000}, {"loss": 0.8943, "grad_norm": 1.7182508707046509, "learning_rate": 0.0002, "epoch": 6.88, "step": 3010}, {"loss": 0.7964, "grad_norm": 1.6308406591415405, "learning_rate": 0.0002, "epoch": 6.902857142857143, "step": 3020}, {"loss": 0.8207, "grad_norm": 1.5080229043960571, "learning_rate": 0.0002, "epoch": 6.925714285714285, "step": 3030}, {"loss": 0.886, "grad_norm": 1.623555064201355, "learning_rate": 0.0002, "epoch": 6.948571428571428, "step": 3040}, {"loss": 0.8377, "grad_norm": 1.526054859161377, "learning_rate": 0.0002, "epoch": 6.9714285714285715, "step": 3050}, {"loss": 0.8816, "grad_norm": 1.6671174764633179, "learning_rate": 0.0002, "epoch": 6.994285714285715, "step": 3060}, {"eval_loss": 2.647613525390625, "eval_runtime": 111.2255, "eval_samples_per_second": 4.558, "eval_steps_per_second": 0.575, "epoch": 6.998857142857143, "step": 3062}, {"loss": 0.6303, "grad_norm": 1.9154540300369263, "learning_rate": 0.0002, "epoch": 7.017142857142857, "step": 3070}, {"loss": 0.5254, "grad_norm": 2.1938717365264893, "learning_rate": 0.0002, "epoch": 7.04, "step": 3080}, {"loss": 0.5087, "grad_norm": 1.7861053943634033, "learning_rate": 0.0002, "epoch": 7.062857142857143, "step": 3090}, {"loss": 0.5751, "grad_norm": 2.096458911895752, "learning_rate": 0.0002, "epoch": 7.085714285714285, "step": 3100}, {"loss": 0.5188, "grad_norm": 2.0057616233825684, "learning_rate": 0.0002, "epoch": 7.1085714285714285, "step": 3110}, {"loss": 0.5544, "grad_norm": 1.7073354721069336, "learning_rate": 0.0002, "epoch": 7.131428571428572, "step": 3120}, {"loss": 0.6005, "grad_norm": 2.3477938175201416, "learning_rate": 0.0002, "epoch": 7.154285714285714, "step": 3130}, {"loss": 0.5303, "grad_norm": 2.0903899669647217, "learning_rate": 0.0002, "epoch": 7.177142857142857, "step": 3140}, {"loss": 0.5397, "grad_norm": 1.7363157272338867, "learning_rate": 0.0002, "epoch": 7.2, "step": 3150}, {"loss": 0.5491, "grad_norm": 2.0611023902893066, "learning_rate": 0.0002, "epoch": 7.222857142857142, "step": 3160}, {"loss": 0.5865, "grad_norm": 2.404407501220703, "learning_rate": 0.0002, "epoch": 7.2457142857142856, "step": 3170}, {"loss": 0.5743, "grad_norm": 2.1841039657592773, "learning_rate": 0.0002, "epoch": 7.268571428571429, "step": 3180}, {"loss": 0.6604, "grad_norm": 1.7582741975784302, "learning_rate": 0.0002, "epoch": 7.291428571428572, "step": 3190}, {"loss": 0.5697, "grad_norm": 1.8890602588653564, "learning_rate": 0.0002, "epoch": 7.314285714285714, "step": 3200}, {"loss": 0.5969, "grad_norm": 1.8433198928833008, "learning_rate": 0.0002, "epoch": 7.337142857142857, "step": 3210}, {"loss": 0.6019, "grad_norm": 1.652266263961792, "learning_rate": 0.0002, "epoch": 7.36, "step": 3220}, {"loss": 0.5439, "grad_norm": 1.914348840713501, "learning_rate": 0.0002, "epoch": 7.382857142857143, "step": 3230}, {"loss": 0.6198, "grad_norm": 1.7440582513809204, "learning_rate": 0.0002, "epoch": 7.405714285714286, "step": 3240}, {"loss": 0.6022, "grad_norm": 1.9745666980743408, "learning_rate": 0.0002, "epoch": 7.428571428571429, "step": 3250}, {"loss": 0.5512, "grad_norm": 1.6567715406417847, "learning_rate": 0.0002, "epoch": 7.451428571428571, "step": 3260}, {"loss": 0.5985, "grad_norm": 1.5239425897598267, "learning_rate": 0.0002, "epoch": 7.474285714285714, "step": 3270}, {"loss": 0.5944, "grad_norm": 2.0668740272521973, "learning_rate": 0.0002, "epoch": 7.497142857142857, "step": 3280}, {"loss": 0.5424, "grad_norm": 1.9551687240600586, "learning_rate": 0.0002, "epoch": 7.52, "step": 3290}, {"loss": 0.6521, "grad_norm": 2.276602268218994, "learning_rate": 0.0002, "epoch": 7.542857142857143, "step": 3300}, {"loss": 0.6183, "grad_norm": 1.9060227870941162, "learning_rate": 0.0002, "epoch": 7.565714285714286, "step": 3310}, {"loss": 0.6151, "grad_norm": 2.0276358127593994, "learning_rate": 0.0002, "epoch": 7.588571428571429, "step": 3320}, {"loss": 0.5731, "grad_norm": 2.037238121032715, "learning_rate": 0.0002, "epoch": 7.611428571428571, "step": 3330}, {"loss": 0.6522, "grad_norm": 2.0060055255889893, "learning_rate": 0.0002, "epoch": 7.634285714285714, "step": 3340}, {"loss": 0.6275, "grad_norm": 1.8366512060165405, "learning_rate": 0.0002, "epoch": 7.6571428571428575, "step": 3350}, {"loss": 0.6535, "grad_norm": 2.0789284706115723, "learning_rate": 0.0002, "epoch": 7.68, "step": 3360}, {"loss": 0.6197, "grad_norm": 2.137089490890503, "learning_rate": 0.0002, "epoch": 7.702857142857143, "step": 3370}, {"loss": 0.6267, "grad_norm": 1.829277753829956, "learning_rate": 0.0002, "epoch": 7.725714285714286, "step": 3380}, {"loss": 0.6567, "grad_norm": 1.9483778476715088, "learning_rate": 0.0002, "epoch": 7.748571428571428, "step": 3390}, {"loss": 0.6393, "grad_norm": 2.0347065925598145, "learning_rate": 0.0002, "epoch": 7.771428571428571, "step": 3400}, {"loss": 0.662, "grad_norm": 2.0142312049865723, "learning_rate": 0.0002, "epoch": 7.7942857142857145, "step": 3410}, {"loss": 0.6349, "grad_norm": 2.152569055557251, "learning_rate": 0.0002, "epoch": 7.817142857142857, "step": 3420}, {"loss": 0.6805, "grad_norm": 1.7300190925598145, "learning_rate": 0.0002, "epoch": 7.84, "step": 3430}, {"loss": 0.6538, "grad_norm": 2.3944954872131348, "learning_rate": 0.0002, "epoch": 7.862857142857143, "step": 3440}, {"loss": 0.6187, "grad_norm": 2.1004269123077393, "learning_rate": 0.0002, "epoch": 7.885714285714286, "step": 3450}, {"loss": 0.6626, "grad_norm": 2.05513072013855, "learning_rate": 0.0002, "epoch": 7.908571428571428, "step": 3460}, {"loss": 0.637, "grad_norm": 1.9822633266448975, "learning_rate": 0.0002, "epoch": 7.9314285714285715, "step": 3470}, {"loss": 0.6663, "grad_norm": 1.9649063348770142, "learning_rate": 0.0002, "epoch": 7.954285714285715, "step": 3480}, {"loss": 0.6192, "grad_norm": 1.7002657651901245, "learning_rate": 0.0002, "epoch": 7.977142857142857, "step": 3490}]}