diff --git a/.gitattributes b/.gitattributes
index 8ae69a34631e9707590877326f7b10eb92b77fff..35d8c15d98613015b1d6248be28832d8a5c1edeb 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -2362,3 +2362,12 @@ gemma-2-9b-it_int4_flare-en-fpb_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1
 gemma-2-9b-it_int4_flare-en-fpb_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-1860-sd-42/checkpoint-812/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_flare-en-fpb_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-1860-sd-42/checkpoint-928/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_flare-en-fpb_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-1860-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5cde1a718c487f3aed9dc296351828322488db0b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79a867fb634fd8e3b066d1e5dba2fcd82385843220a23a334f7884c4c27c8b96
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5cde1a718c487f3aed9dc296351828322488db0b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79a867fb634fd8e3b066d1e5dba2fcd82385843220a23a334f7884c4c27c8b96
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e3939ef6796641269f6ace9087d94bbd3646c41a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4c65999d237dae4424b1944a70de5680c46f42713e3f2c8537a7c2ab85c5104
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1d1cb00e182f171d46f445cd099268e812deda8e
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49e211764da55bdbda0c95564a513ca390fea4afa7b89981b7c84330cfeb75ce
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c3d5fc807a13a3f6eccccd8ac1c5a837c97f3ef2
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38841519e925109736ce493a81a32baf57374f07cc4493ecbc110955926f5fa2
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..dd4f57b1018a848506eaaaf6dc64a8c5cadaffce
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/trainer_state.json
@@ -0,0 +1,1323 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 1827,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 9.384156326264832e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6f2d144ec6a452a0f1a79ed3c4f07edef4497c12
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:035921963b7ba3426c7b25b29c3e4e62d3764d0ef3b24abb0b53606c314dc350
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..444b1b7b10d84d1c897d3a25b29cd96d5c716d38
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d210241b2fbe25b1e819460216e4143f1484c8e6fc3636c95d1498e9b095ca51
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1acbae1b01538922e3bad83dee13f59af4330f7e
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf917d2e89ee39a3e9cdfb5c9be783c7806b020f29415fe37710fbbf43833346
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7f0dc91e31a4046330b836b2990c20d6b3f2f25e
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c054e94e5454455e09f8cd2be3390222c38c11f62f45d2b0a3f8280bdbc1c254
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9a29bf43ac8715abdcc5aa8f9336e317817624c7
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/trainer_state.json
@@ -0,0 +1,1975 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 2.9994526546250686,
+  "eval_steps": 10,
+  "global_step": 2740,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 1.4076234489397248e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-2740/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b37c20bd19b840d16f58d759d64f28458e6c763c
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a84388d4ac5738c377bd47af9ca50506439a278bc680479cf2db90b15345be18
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4f9b81e2e43136a24c63b7a88ec19f174e1c3141
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93866aa1024d72e1ac54b17d93e37aa192ee0d448c3afaeafe3afb9d48dbf0de
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d0520f243b89737759e558104789698fffbe6097
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b06f9b81fb40fc97b17f99738dd19e250800243191cab070d3f261467f044a84
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..07e4755f3a4387c1a7422a665bea15f78d9493f7
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5002c510756345ccba22822baa17f31534958c278f8a15afb59f45002a8e21d9
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7a222fa137afff280b5bdc15dcccc9b1ec90571a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/trainer_state.json
@@ -0,0 +1,2620 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 3654,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    },
+    {
+      "epoch": 3.0103995621237,
+      "grad_norm": 0.6384502053260803,
+      "learning_rate": 0.0002,
+      "loss": 0.6443,
+      "step": 2750
+    },
+    {
+      "epoch": 3.0213464696223316,
+      "grad_norm": 0.9928722381591797,
+      "learning_rate": 0.0002,
+      "loss": 0.7123,
+      "step": 2760
+    },
+    {
+      "epoch": 3.0322933771209635,
+      "grad_norm": 0.7813051342964172,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 2770
+    },
+    {
+      "epoch": 3.043240284619595,
+      "grad_norm": 1.0202556848526,
+      "learning_rate": 0.0002,
+      "loss": 0.6042,
+      "step": 2780
+    },
+    {
+      "epoch": 3.0541871921182264,
+      "grad_norm": 0.7581062316894531,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 2790
+    },
+    {
+      "epoch": 3.0651340996168583,
+      "grad_norm": 0.6252710223197937,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 2800
+    },
+    {
+      "epoch": 3.07608100711549,
+      "grad_norm": 0.7738662958145142,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 2810
+    },
+    {
+      "epoch": 3.0870279146141213,
+      "grad_norm": 0.7381885051727295,
+      "learning_rate": 0.0002,
+      "loss": 0.627,
+      "step": 2820
+    },
+    {
+      "epoch": 3.097974822112753,
+      "grad_norm": 0.9197564721107483,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 2830
+    },
+    {
+      "epoch": 3.1089217296113847,
+      "grad_norm": 1.000976800918579,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 2840
+    },
+    {
+      "epoch": 3.1198686371100166,
+      "grad_norm": 0.7559131383895874,
+      "learning_rate": 0.0002,
+      "loss": 0.6631,
+      "step": 2850
+    },
+    {
+      "epoch": 3.130815544608648,
+      "grad_norm": 0.7213780879974365,
+      "learning_rate": 0.0002,
+      "loss": 0.6252,
+      "step": 2860
+    },
+    {
+      "epoch": 3.1417624521072796,
+      "grad_norm": 0.945939838886261,
+      "learning_rate": 0.0002,
+      "loss": 0.6501,
+      "step": 2870
+    },
+    {
+      "epoch": 3.1527093596059115,
+      "grad_norm": 0.7277454137802124,
+      "learning_rate": 0.0002,
+      "loss": 0.6129,
+      "step": 2880
+    },
+    {
+      "epoch": 3.163656267104543,
+      "grad_norm": 0.762026846408844,
+      "learning_rate": 0.0002,
+      "loss": 0.6423,
+      "step": 2890
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6471221446990967,
+      "learning_rate": 0.0002,
+      "loss": 0.5332,
+      "step": 2900
+    },
+    {
+      "epoch": 3.1855500821018063,
+      "grad_norm": 0.6018978357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 2910
+    },
+    {
+      "epoch": 3.196496989600438,
+      "grad_norm": 0.8607320785522461,
+      "learning_rate": 0.0002,
+      "loss": 0.7274,
+      "step": 2920
+    },
+    {
+      "epoch": 3.2074438970990693,
+      "grad_norm": 0.8854126334190369,
+      "learning_rate": 0.0002,
+      "loss": 0.6139,
+      "step": 2930
+    },
+    {
+      "epoch": 3.218390804597701,
+      "grad_norm": 0.6620870232582092,
+      "learning_rate": 0.0002,
+      "loss": 0.6485,
+      "step": 2940
+    },
+    {
+      "epoch": 3.2293377120963327,
+      "grad_norm": 0.7377511858940125,
+      "learning_rate": 0.0002,
+      "loss": 0.6969,
+      "step": 2950
+    },
+    {
+      "epoch": 3.2402846195949646,
+      "grad_norm": 0.7803301811218262,
+      "learning_rate": 0.0002,
+      "loss": 0.6798,
+      "step": 2960
+    },
+    {
+      "epoch": 3.251231527093596,
+      "grad_norm": 0.834061861038208,
+      "learning_rate": 0.0002,
+      "loss": 0.6697,
+      "step": 2970
+    },
+    {
+      "epoch": 3.2621784345922276,
+      "grad_norm": 0.8496041893959045,
+      "learning_rate": 0.0002,
+      "loss": 0.6894,
+      "step": 2980
+    },
+    {
+      "epoch": 3.2731253420908595,
+      "grad_norm": 0.7967984676361084,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 2990
+    },
+    {
+      "epoch": 3.284072249589491,
+      "grad_norm": 1.0207016468048096,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 3000
+    },
+    {
+      "epoch": 3.2950191570881224,
+      "grad_norm": 0.7855771780014038,
+      "learning_rate": 0.0002,
+      "loss": 0.6586,
+      "step": 3010
+    },
+    {
+      "epoch": 3.3059660645867543,
+      "grad_norm": 0.9427546858787537,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 3020
+    },
+    {
+      "epoch": 3.316912972085386,
+      "grad_norm": 0.823542594909668,
+      "learning_rate": 0.0002,
+      "loss": 0.6277,
+      "step": 3030
+    },
+    {
+      "epoch": 3.3278598795840173,
+      "grad_norm": 0.9826635122299194,
+      "learning_rate": 0.0002,
+      "loss": 0.7109,
+      "step": 3040
+    },
+    {
+      "epoch": 3.338806787082649,
+      "grad_norm": 0.7259827852249146,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 3050
+    },
+    {
+      "epoch": 3.3497536945812807,
+      "grad_norm": 0.7774739861488342,
+      "learning_rate": 0.0002,
+      "loss": 0.653,
+      "step": 3060
+    },
+    {
+      "epoch": 3.3607006020799126,
+      "grad_norm": 0.7394293546676636,
+      "learning_rate": 0.0002,
+      "loss": 0.7529,
+      "step": 3070
+    },
+    {
+      "epoch": 3.371647509578544,
+      "grad_norm": 0.9017578959465027,
+      "learning_rate": 0.0002,
+      "loss": 0.5987,
+      "step": 3080
+    },
+    {
+      "epoch": 3.3825944170771756,
+      "grad_norm": 0.7451054453849792,
+      "learning_rate": 0.0002,
+      "loss": 0.6953,
+      "step": 3090
+    },
+    {
+      "epoch": 3.3935413245758075,
+      "grad_norm": 0.7321506142616272,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 3100
+    },
+    {
+      "epoch": 3.404488232074439,
+      "grad_norm": 0.6721828579902649,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3110
+    },
+    {
+      "epoch": 3.4154351395730704,
+      "grad_norm": 0.774022102355957,
+      "learning_rate": 0.0002,
+      "loss": 0.6559,
+      "step": 3120
+    },
+    {
+      "epoch": 3.4263820470717024,
+      "grad_norm": 0.9143537282943726,
+      "learning_rate": 0.0002,
+      "loss": 0.7449,
+      "step": 3130
+    },
+    {
+      "epoch": 3.437328954570334,
+      "grad_norm": 1.226087212562561,
+      "learning_rate": 0.0002,
+      "loss": 0.6899,
+      "step": 3140
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7545496225357056,
+      "learning_rate": 0.0002,
+      "loss": 0.6719,
+      "step": 3150
+    },
+    {
+      "epoch": 3.4592227695675972,
+      "grad_norm": 0.6515635848045349,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 3160
+    },
+    {
+      "epoch": 3.4701696770662287,
+      "grad_norm": 0.9297090172767639,
+      "learning_rate": 0.0002,
+      "loss": 0.6926,
+      "step": 3170
+    },
+    {
+      "epoch": 3.4811165845648606,
+      "grad_norm": 1.0130730867385864,
+      "learning_rate": 0.0002,
+      "loss": 0.6071,
+      "step": 3180
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.7654589414596558,
+      "learning_rate": 0.0002,
+      "loss": 0.5959,
+      "step": 3190
+    },
+    {
+      "epoch": 3.5030103995621236,
+      "grad_norm": 0.9954977631568909,
+      "learning_rate": 0.0002,
+      "loss": 0.7401,
+      "step": 3200
+    },
+    {
+      "epoch": 3.5139573070607555,
+      "grad_norm": 0.6027487516403198,
+      "learning_rate": 0.0002,
+      "loss": 0.6661,
+      "step": 3210
+    },
+    {
+      "epoch": 3.524904214559387,
+      "grad_norm": 0.741770327091217,
+      "learning_rate": 0.0002,
+      "loss": 0.6963,
+      "step": 3220
+    },
+    {
+      "epoch": 3.535851122058019,
+      "grad_norm": 1.0534909963607788,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 3230
+    },
+    {
+      "epoch": 3.5467980295566504,
+      "grad_norm": 0.937772274017334,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 3240
+    },
+    {
+      "epoch": 3.557744937055282,
+      "grad_norm": 0.8504213690757751,
+      "learning_rate": 0.0002,
+      "loss": 0.6681,
+      "step": 3250
+    },
+    {
+      "epoch": 3.5686918445539133,
+      "grad_norm": 0.7755007147789001,
+      "learning_rate": 0.0002,
+      "loss": 0.6436,
+      "step": 3260
+    },
+    {
+      "epoch": 3.5796387520525452,
+      "grad_norm": 1.0193358659744263,
+      "learning_rate": 0.0002,
+      "loss": 0.6213,
+      "step": 3270
+    },
+    {
+      "epoch": 3.5905856595511767,
+      "grad_norm": 0.8440536856651306,
+      "learning_rate": 0.0002,
+      "loss": 0.671,
+      "step": 3280
+    },
+    {
+      "epoch": 3.6015325670498086,
+      "grad_norm": 0.6195939183235168,
+      "learning_rate": 0.0002,
+      "loss": 0.6859,
+      "step": 3290
+    },
+    {
+      "epoch": 3.61247947454844,
+      "grad_norm": 0.8608590960502625,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3300
+    },
+    {
+      "epoch": 3.6234263820470716,
+      "grad_norm": 0.6772327423095703,
+      "learning_rate": 0.0002,
+      "loss": 0.7301,
+      "step": 3310
+    },
+    {
+      "epoch": 3.6343732895457035,
+      "grad_norm": 0.8031839728355408,
+      "learning_rate": 0.0002,
+      "loss": 0.6298,
+      "step": 3320
+    },
+    {
+      "epoch": 3.645320197044335,
+      "grad_norm": 0.6080502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.7041,
+      "step": 3330
+    },
+    {
+      "epoch": 3.656267104542967,
+      "grad_norm": 0.8007240891456604,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 3340
+    },
+    {
+      "epoch": 3.6672140120415984,
+      "grad_norm": 0.8060704469680786,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3350
+    },
+    {
+      "epoch": 3.67816091954023,
+      "grad_norm": 0.7547586560249329,
+      "learning_rate": 0.0002,
+      "loss": 0.6304,
+      "step": 3360
+    },
+    {
+      "epoch": 3.6891078270388613,
+      "grad_norm": 0.686851978302002,
+      "learning_rate": 0.0002,
+      "loss": 0.7066,
+      "step": 3370
+    },
+    {
+      "epoch": 3.7000547345374932,
+      "grad_norm": 0.9429075717926025,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 3380
+    },
+    {
+      "epoch": 3.7110016420361247,
+      "grad_norm": 0.7283591032028198,
+      "learning_rate": 0.0002,
+      "loss": 0.6673,
+      "step": 3390
+    },
+    {
+      "epoch": 3.7219485495347566,
+      "grad_norm": 0.8323085904121399,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 3400
+    },
+    {
+      "epoch": 3.732895457033388,
+      "grad_norm": 0.8529590964317322,
+      "learning_rate": 0.0002,
+      "loss": 0.7779,
+      "step": 3410
+    },
+    {
+      "epoch": 3.7438423645320196,
+      "grad_norm": 0.731752872467041,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3420
+    },
+    {
+      "epoch": 3.7547892720306515,
+      "grad_norm": 0.8572278618812561,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3430
+    },
+    {
+      "epoch": 3.765736179529283,
+      "grad_norm": 0.7408691048622131,
+      "learning_rate": 0.0002,
+      "loss": 0.6215,
+      "step": 3440
+    },
+    {
+      "epoch": 3.776683087027915,
+      "grad_norm": 0.7470445036888123,
+      "learning_rate": 0.0002,
+      "loss": 0.622,
+      "step": 3450
+    },
+    {
+      "epoch": 3.7876299945265464,
+      "grad_norm": 0.6806244254112244,
+      "learning_rate": 0.0002,
+      "loss": 0.7241,
+      "step": 3460
+    },
+    {
+      "epoch": 3.798576902025178,
+      "grad_norm": 0.9129069447517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7739,
+      "step": 3470
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.8717501759529114,
+      "learning_rate": 0.0002,
+      "loss": 0.6826,
+      "step": 3480
+    },
+    {
+      "epoch": 3.8204707170224412,
+      "grad_norm": 0.6761979460716248,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 3490
+    },
+    {
+      "epoch": 3.8314176245210727,
+      "grad_norm": 1.0054380893707275,
+      "learning_rate": 0.0002,
+      "loss": 0.7601,
+      "step": 3500
+    },
+    {
+      "epoch": 3.8423645320197046,
+      "grad_norm": 1.1224009990692139,
+      "learning_rate": 0.0002,
+      "loss": 0.623,
+      "step": 3510
+    },
+    {
+      "epoch": 3.853311439518336,
+      "grad_norm": 0.8997692465782166,
+      "learning_rate": 0.0002,
+      "loss": 0.6918,
+      "step": 3520
+    },
+    {
+      "epoch": 3.8642583470169676,
+      "grad_norm": 1.0086902379989624,
+      "learning_rate": 0.0002,
+      "loss": 0.6357,
+      "step": 3530
+    },
+    {
+      "epoch": 3.8752052545155995,
+      "grad_norm": 0.772739589214325,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 3540
+    },
+    {
+      "epoch": 3.886152162014231,
+      "grad_norm": 1.211774230003357,
+      "learning_rate": 0.0002,
+      "loss": 0.7423,
+      "step": 3550
+    },
+    {
+      "epoch": 3.897099069512863,
+      "grad_norm": 0.9572356939315796,
+      "learning_rate": 0.0002,
+      "loss": 0.7321,
+      "step": 3560
+    },
+    {
+      "epoch": 3.9080459770114944,
+      "grad_norm": 0.7887842655181885,
+      "learning_rate": 0.0002,
+      "loss": 0.6836,
+      "step": 3570
+    },
+    {
+      "epoch": 3.918992884510126,
+      "grad_norm": 0.7308389544487,
+      "learning_rate": 0.0002,
+      "loss": 0.7576,
+      "step": 3580
+    },
+    {
+      "epoch": 3.9299397920087573,
+      "grad_norm": 1.0182650089263916,
+      "learning_rate": 0.0002,
+      "loss": 0.6001,
+      "step": 3590
+    },
+    {
+      "epoch": 3.9408866995073892,
+      "grad_norm": 0.8000147342681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 3600
+    },
+    {
+      "epoch": 3.9518336070060207,
+      "grad_norm": 0.7385728359222412,
+      "learning_rate": 0.0002,
+      "loss": 0.6244,
+      "step": 3610
+    },
+    {
+      "epoch": 3.9627805145046526,
+      "grad_norm": 0.9233261942863464,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 3620
+    },
+    {
+      "epoch": 3.973727422003284,
+      "grad_norm": 0.8486751914024353,
+      "learning_rate": 0.0002,
+      "loss": 0.6508,
+      "step": 3630
+    },
+    {
+      "epoch": 3.9846743295019156,
+      "grad_norm": 0.7593663334846497,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3640
+    },
+    {
+      "epoch": 3.9956212370005475,
+      "grad_norm": 0.7885415554046631,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 3650
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.250312328338623,
+      "eval_runtime": 46.0842,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.193,
+      "step": 3654
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 1.8768312652529664e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-3654/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..54340503512ec66744b9f6346df09f21a03940b3
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:690f0a108c96b12c41c037e7a8a073b10e97220dd5bf277ebaf85d1405a0e83a
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3ee5cacec4cdd22a3720a8050aa34b695971059b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a60d365e1ec200496f2cba711983f8608c9cb7d697c7c2b44a16f4348095ae96
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..60d7a14ea93086849c11aba92a635f5276eb71eb
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ec7f713eac7d32af496212fbeebae513684ce562bcd5df63eb5603a6c82d35d
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8b92c057ff7f770e9f74e39b0230dd2919efdd5b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81183db37481adb9d2771807b01e740317fdbbc85a4be1d990e73274e3f03be7
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..4c1d3ca3731f68e72003ae018a16d840358b0a8d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/trainer_state.json
@@ -0,0 +1,3265 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 4.999452654625069,
+  "eval_steps": 10,
+  "global_step": 4567,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    },
+    {
+      "epoch": 3.0103995621237,
+      "grad_norm": 0.6384502053260803,
+      "learning_rate": 0.0002,
+      "loss": 0.6443,
+      "step": 2750
+    },
+    {
+      "epoch": 3.0213464696223316,
+      "grad_norm": 0.9928722381591797,
+      "learning_rate": 0.0002,
+      "loss": 0.7123,
+      "step": 2760
+    },
+    {
+      "epoch": 3.0322933771209635,
+      "grad_norm": 0.7813051342964172,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 2770
+    },
+    {
+      "epoch": 3.043240284619595,
+      "grad_norm": 1.0202556848526,
+      "learning_rate": 0.0002,
+      "loss": 0.6042,
+      "step": 2780
+    },
+    {
+      "epoch": 3.0541871921182264,
+      "grad_norm": 0.7581062316894531,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 2790
+    },
+    {
+      "epoch": 3.0651340996168583,
+      "grad_norm": 0.6252710223197937,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 2800
+    },
+    {
+      "epoch": 3.07608100711549,
+      "grad_norm": 0.7738662958145142,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 2810
+    },
+    {
+      "epoch": 3.0870279146141213,
+      "grad_norm": 0.7381885051727295,
+      "learning_rate": 0.0002,
+      "loss": 0.627,
+      "step": 2820
+    },
+    {
+      "epoch": 3.097974822112753,
+      "grad_norm": 0.9197564721107483,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 2830
+    },
+    {
+      "epoch": 3.1089217296113847,
+      "grad_norm": 1.000976800918579,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 2840
+    },
+    {
+      "epoch": 3.1198686371100166,
+      "grad_norm": 0.7559131383895874,
+      "learning_rate": 0.0002,
+      "loss": 0.6631,
+      "step": 2850
+    },
+    {
+      "epoch": 3.130815544608648,
+      "grad_norm": 0.7213780879974365,
+      "learning_rate": 0.0002,
+      "loss": 0.6252,
+      "step": 2860
+    },
+    {
+      "epoch": 3.1417624521072796,
+      "grad_norm": 0.945939838886261,
+      "learning_rate": 0.0002,
+      "loss": 0.6501,
+      "step": 2870
+    },
+    {
+      "epoch": 3.1527093596059115,
+      "grad_norm": 0.7277454137802124,
+      "learning_rate": 0.0002,
+      "loss": 0.6129,
+      "step": 2880
+    },
+    {
+      "epoch": 3.163656267104543,
+      "grad_norm": 0.762026846408844,
+      "learning_rate": 0.0002,
+      "loss": 0.6423,
+      "step": 2890
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6471221446990967,
+      "learning_rate": 0.0002,
+      "loss": 0.5332,
+      "step": 2900
+    },
+    {
+      "epoch": 3.1855500821018063,
+      "grad_norm": 0.6018978357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 2910
+    },
+    {
+      "epoch": 3.196496989600438,
+      "grad_norm": 0.8607320785522461,
+      "learning_rate": 0.0002,
+      "loss": 0.7274,
+      "step": 2920
+    },
+    {
+      "epoch": 3.2074438970990693,
+      "grad_norm": 0.8854126334190369,
+      "learning_rate": 0.0002,
+      "loss": 0.6139,
+      "step": 2930
+    },
+    {
+      "epoch": 3.218390804597701,
+      "grad_norm": 0.6620870232582092,
+      "learning_rate": 0.0002,
+      "loss": 0.6485,
+      "step": 2940
+    },
+    {
+      "epoch": 3.2293377120963327,
+      "grad_norm": 0.7377511858940125,
+      "learning_rate": 0.0002,
+      "loss": 0.6969,
+      "step": 2950
+    },
+    {
+      "epoch": 3.2402846195949646,
+      "grad_norm": 0.7803301811218262,
+      "learning_rate": 0.0002,
+      "loss": 0.6798,
+      "step": 2960
+    },
+    {
+      "epoch": 3.251231527093596,
+      "grad_norm": 0.834061861038208,
+      "learning_rate": 0.0002,
+      "loss": 0.6697,
+      "step": 2970
+    },
+    {
+      "epoch": 3.2621784345922276,
+      "grad_norm": 0.8496041893959045,
+      "learning_rate": 0.0002,
+      "loss": 0.6894,
+      "step": 2980
+    },
+    {
+      "epoch": 3.2731253420908595,
+      "grad_norm": 0.7967984676361084,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 2990
+    },
+    {
+      "epoch": 3.284072249589491,
+      "grad_norm": 1.0207016468048096,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 3000
+    },
+    {
+      "epoch": 3.2950191570881224,
+      "grad_norm": 0.7855771780014038,
+      "learning_rate": 0.0002,
+      "loss": 0.6586,
+      "step": 3010
+    },
+    {
+      "epoch": 3.3059660645867543,
+      "grad_norm": 0.9427546858787537,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 3020
+    },
+    {
+      "epoch": 3.316912972085386,
+      "grad_norm": 0.823542594909668,
+      "learning_rate": 0.0002,
+      "loss": 0.6277,
+      "step": 3030
+    },
+    {
+      "epoch": 3.3278598795840173,
+      "grad_norm": 0.9826635122299194,
+      "learning_rate": 0.0002,
+      "loss": 0.7109,
+      "step": 3040
+    },
+    {
+      "epoch": 3.338806787082649,
+      "grad_norm": 0.7259827852249146,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 3050
+    },
+    {
+      "epoch": 3.3497536945812807,
+      "grad_norm": 0.7774739861488342,
+      "learning_rate": 0.0002,
+      "loss": 0.653,
+      "step": 3060
+    },
+    {
+      "epoch": 3.3607006020799126,
+      "grad_norm": 0.7394293546676636,
+      "learning_rate": 0.0002,
+      "loss": 0.7529,
+      "step": 3070
+    },
+    {
+      "epoch": 3.371647509578544,
+      "grad_norm": 0.9017578959465027,
+      "learning_rate": 0.0002,
+      "loss": 0.5987,
+      "step": 3080
+    },
+    {
+      "epoch": 3.3825944170771756,
+      "grad_norm": 0.7451054453849792,
+      "learning_rate": 0.0002,
+      "loss": 0.6953,
+      "step": 3090
+    },
+    {
+      "epoch": 3.3935413245758075,
+      "grad_norm": 0.7321506142616272,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 3100
+    },
+    {
+      "epoch": 3.404488232074439,
+      "grad_norm": 0.6721828579902649,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3110
+    },
+    {
+      "epoch": 3.4154351395730704,
+      "grad_norm": 0.774022102355957,
+      "learning_rate": 0.0002,
+      "loss": 0.6559,
+      "step": 3120
+    },
+    {
+      "epoch": 3.4263820470717024,
+      "grad_norm": 0.9143537282943726,
+      "learning_rate": 0.0002,
+      "loss": 0.7449,
+      "step": 3130
+    },
+    {
+      "epoch": 3.437328954570334,
+      "grad_norm": 1.226087212562561,
+      "learning_rate": 0.0002,
+      "loss": 0.6899,
+      "step": 3140
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7545496225357056,
+      "learning_rate": 0.0002,
+      "loss": 0.6719,
+      "step": 3150
+    },
+    {
+      "epoch": 3.4592227695675972,
+      "grad_norm": 0.6515635848045349,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 3160
+    },
+    {
+      "epoch": 3.4701696770662287,
+      "grad_norm": 0.9297090172767639,
+      "learning_rate": 0.0002,
+      "loss": 0.6926,
+      "step": 3170
+    },
+    {
+      "epoch": 3.4811165845648606,
+      "grad_norm": 1.0130730867385864,
+      "learning_rate": 0.0002,
+      "loss": 0.6071,
+      "step": 3180
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.7654589414596558,
+      "learning_rate": 0.0002,
+      "loss": 0.5959,
+      "step": 3190
+    },
+    {
+      "epoch": 3.5030103995621236,
+      "grad_norm": 0.9954977631568909,
+      "learning_rate": 0.0002,
+      "loss": 0.7401,
+      "step": 3200
+    },
+    {
+      "epoch": 3.5139573070607555,
+      "grad_norm": 0.6027487516403198,
+      "learning_rate": 0.0002,
+      "loss": 0.6661,
+      "step": 3210
+    },
+    {
+      "epoch": 3.524904214559387,
+      "grad_norm": 0.741770327091217,
+      "learning_rate": 0.0002,
+      "loss": 0.6963,
+      "step": 3220
+    },
+    {
+      "epoch": 3.535851122058019,
+      "grad_norm": 1.0534909963607788,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 3230
+    },
+    {
+      "epoch": 3.5467980295566504,
+      "grad_norm": 0.937772274017334,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 3240
+    },
+    {
+      "epoch": 3.557744937055282,
+      "grad_norm": 0.8504213690757751,
+      "learning_rate": 0.0002,
+      "loss": 0.6681,
+      "step": 3250
+    },
+    {
+      "epoch": 3.5686918445539133,
+      "grad_norm": 0.7755007147789001,
+      "learning_rate": 0.0002,
+      "loss": 0.6436,
+      "step": 3260
+    },
+    {
+      "epoch": 3.5796387520525452,
+      "grad_norm": 1.0193358659744263,
+      "learning_rate": 0.0002,
+      "loss": 0.6213,
+      "step": 3270
+    },
+    {
+      "epoch": 3.5905856595511767,
+      "grad_norm": 0.8440536856651306,
+      "learning_rate": 0.0002,
+      "loss": 0.671,
+      "step": 3280
+    },
+    {
+      "epoch": 3.6015325670498086,
+      "grad_norm": 0.6195939183235168,
+      "learning_rate": 0.0002,
+      "loss": 0.6859,
+      "step": 3290
+    },
+    {
+      "epoch": 3.61247947454844,
+      "grad_norm": 0.8608590960502625,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3300
+    },
+    {
+      "epoch": 3.6234263820470716,
+      "grad_norm": 0.6772327423095703,
+      "learning_rate": 0.0002,
+      "loss": 0.7301,
+      "step": 3310
+    },
+    {
+      "epoch": 3.6343732895457035,
+      "grad_norm": 0.8031839728355408,
+      "learning_rate": 0.0002,
+      "loss": 0.6298,
+      "step": 3320
+    },
+    {
+      "epoch": 3.645320197044335,
+      "grad_norm": 0.6080502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.7041,
+      "step": 3330
+    },
+    {
+      "epoch": 3.656267104542967,
+      "grad_norm": 0.8007240891456604,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 3340
+    },
+    {
+      "epoch": 3.6672140120415984,
+      "grad_norm": 0.8060704469680786,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3350
+    },
+    {
+      "epoch": 3.67816091954023,
+      "grad_norm": 0.7547586560249329,
+      "learning_rate": 0.0002,
+      "loss": 0.6304,
+      "step": 3360
+    },
+    {
+      "epoch": 3.6891078270388613,
+      "grad_norm": 0.686851978302002,
+      "learning_rate": 0.0002,
+      "loss": 0.7066,
+      "step": 3370
+    },
+    {
+      "epoch": 3.7000547345374932,
+      "grad_norm": 0.9429075717926025,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 3380
+    },
+    {
+      "epoch": 3.7110016420361247,
+      "grad_norm": 0.7283591032028198,
+      "learning_rate": 0.0002,
+      "loss": 0.6673,
+      "step": 3390
+    },
+    {
+      "epoch": 3.7219485495347566,
+      "grad_norm": 0.8323085904121399,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 3400
+    },
+    {
+      "epoch": 3.732895457033388,
+      "grad_norm": 0.8529590964317322,
+      "learning_rate": 0.0002,
+      "loss": 0.7779,
+      "step": 3410
+    },
+    {
+      "epoch": 3.7438423645320196,
+      "grad_norm": 0.731752872467041,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3420
+    },
+    {
+      "epoch": 3.7547892720306515,
+      "grad_norm": 0.8572278618812561,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3430
+    },
+    {
+      "epoch": 3.765736179529283,
+      "grad_norm": 0.7408691048622131,
+      "learning_rate": 0.0002,
+      "loss": 0.6215,
+      "step": 3440
+    },
+    {
+      "epoch": 3.776683087027915,
+      "grad_norm": 0.7470445036888123,
+      "learning_rate": 0.0002,
+      "loss": 0.622,
+      "step": 3450
+    },
+    {
+      "epoch": 3.7876299945265464,
+      "grad_norm": 0.6806244254112244,
+      "learning_rate": 0.0002,
+      "loss": 0.7241,
+      "step": 3460
+    },
+    {
+      "epoch": 3.798576902025178,
+      "grad_norm": 0.9129069447517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7739,
+      "step": 3470
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.8717501759529114,
+      "learning_rate": 0.0002,
+      "loss": 0.6826,
+      "step": 3480
+    },
+    {
+      "epoch": 3.8204707170224412,
+      "grad_norm": 0.6761979460716248,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 3490
+    },
+    {
+      "epoch": 3.8314176245210727,
+      "grad_norm": 1.0054380893707275,
+      "learning_rate": 0.0002,
+      "loss": 0.7601,
+      "step": 3500
+    },
+    {
+      "epoch": 3.8423645320197046,
+      "grad_norm": 1.1224009990692139,
+      "learning_rate": 0.0002,
+      "loss": 0.623,
+      "step": 3510
+    },
+    {
+      "epoch": 3.853311439518336,
+      "grad_norm": 0.8997692465782166,
+      "learning_rate": 0.0002,
+      "loss": 0.6918,
+      "step": 3520
+    },
+    {
+      "epoch": 3.8642583470169676,
+      "grad_norm": 1.0086902379989624,
+      "learning_rate": 0.0002,
+      "loss": 0.6357,
+      "step": 3530
+    },
+    {
+      "epoch": 3.8752052545155995,
+      "grad_norm": 0.772739589214325,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 3540
+    },
+    {
+      "epoch": 3.886152162014231,
+      "grad_norm": 1.211774230003357,
+      "learning_rate": 0.0002,
+      "loss": 0.7423,
+      "step": 3550
+    },
+    {
+      "epoch": 3.897099069512863,
+      "grad_norm": 0.9572356939315796,
+      "learning_rate": 0.0002,
+      "loss": 0.7321,
+      "step": 3560
+    },
+    {
+      "epoch": 3.9080459770114944,
+      "grad_norm": 0.7887842655181885,
+      "learning_rate": 0.0002,
+      "loss": 0.6836,
+      "step": 3570
+    },
+    {
+      "epoch": 3.918992884510126,
+      "grad_norm": 0.7308389544487,
+      "learning_rate": 0.0002,
+      "loss": 0.7576,
+      "step": 3580
+    },
+    {
+      "epoch": 3.9299397920087573,
+      "grad_norm": 1.0182650089263916,
+      "learning_rate": 0.0002,
+      "loss": 0.6001,
+      "step": 3590
+    },
+    {
+      "epoch": 3.9408866995073892,
+      "grad_norm": 0.8000147342681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 3600
+    },
+    {
+      "epoch": 3.9518336070060207,
+      "grad_norm": 0.7385728359222412,
+      "learning_rate": 0.0002,
+      "loss": 0.6244,
+      "step": 3610
+    },
+    {
+      "epoch": 3.9627805145046526,
+      "grad_norm": 0.9233261942863464,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 3620
+    },
+    {
+      "epoch": 3.973727422003284,
+      "grad_norm": 0.8486751914024353,
+      "learning_rate": 0.0002,
+      "loss": 0.6508,
+      "step": 3630
+    },
+    {
+      "epoch": 3.9846743295019156,
+      "grad_norm": 0.7593663334846497,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3640
+    },
+    {
+      "epoch": 3.9956212370005475,
+      "grad_norm": 0.7885415554046631,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 3650
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.250312328338623,
+      "eval_runtime": 46.0842,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.193,
+      "step": 3654
+    },
+    {
+      "epoch": 4.006568144499179,
+      "grad_norm": 0.6591703295707703,
+      "learning_rate": 0.0002,
+      "loss": 0.5547,
+      "step": 3660
+    },
+    {
+      "epoch": 4.017515051997811,
+      "grad_norm": 1.36927330493927,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 3670
+    },
+    {
+      "epoch": 4.028461959496442,
+      "grad_norm": 0.8106328845024109,
+      "learning_rate": 0.0002,
+      "loss": 0.4466,
+      "step": 3680
+    },
+    {
+      "epoch": 4.039408866995074,
+      "grad_norm": 0.7592712044715881,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 3690
+    },
+    {
+      "epoch": 4.050355774493705,
+      "grad_norm": 0.9518909454345703,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3700
+    },
+    {
+      "epoch": 4.061302681992337,
+      "grad_norm": 0.7805967330932617,
+      "learning_rate": 0.0002,
+      "loss": 0.4638,
+      "step": 3710
+    },
+    {
+      "epoch": 4.072249589490969,
+      "grad_norm": 1.3146334886550903,
+      "learning_rate": 0.0002,
+      "loss": 0.4556,
+      "step": 3720
+    },
+    {
+      "epoch": 4.083196496989601,
+      "grad_norm": 1.1611138582229614,
+      "learning_rate": 0.0002,
+      "loss": 0.5635,
+      "step": 3730
+    },
+    {
+      "epoch": 4.094143404488232,
+      "grad_norm": 0.8173232078552246,
+      "learning_rate": 0.0002,
+      "loss": 0.3845,
+      "step": 3740
+    },
+    {
+      "epoch": 4.105090311986864,
+      "grad_norm": 0.7848323583602905,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 3750
+    },
+    {
+      "epoch": 4.116037219485495,
+      "grad_norm": 1.3183201551437378,
+      "learning_rate": 0.0002,
+      "loss": 0.4519,
+      "step": 3760
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 1.1936529874801636,
+      "learning_rate": 0.0002,
+      "loss": 0.5083,
+      "step": 3770
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1078993082046509,
+      "learning_rate": 0.0002,
+      "loss": 0.5208,
+      "step": 3780
+    },
+    {
+      "epoch": 4.14887794198139,
+      "grad_norm": 1.107743263244629,
+      "learning_rate": 0.0002,
+      "loss": 0.5928,
+      "step": 3790
+    },
+    {
+      "epoch": 4.159824849480022,
+      "grad_norm": 0.7801875472068787,
+      "learning_rate": 0.0002,
+      "loss": 0.5112,
+      "step": 3800
+    },
+    {
+      "epoch": 4.170771756978653,
+      "grad_norm": 1.1328117847442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4896,
+      "step": 3810
+    },
+    {
+      "epoch": 4.181718664477285,
+      "grad_norm": 1.4232193231582642,
+      "learning_rate": 0.0002,
+      "loss": 0.5645,
+      "step": 3820
+    },
+    {
+      "epoch": 4.192665571975917,
+      "grad_norm": 1.557416558265686,
+      "learning_rate": 0.0002,
+      "loss": 0.5049,
+      "step": 3830
+    },
+    {
+      "epoch": 4.203612479474549,
+      "grad_norm": 1.042923092842102,
+      "learning_rate": 0.0002,
+      "loss": 0.4863,
+      "step": 3840
+    },
+    {
+      "epoch": 4.21455938697318,
+      "grad_norm": 1.1801949739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 3850
+    },
+    {
+      "epoch": 4.225506294471812,
+      "grad_norm": 0.9273753762245178,
+      "learning_rate": 0.0002,
+      "loss": 0.5063,
+      "step": 3860
+    },
+    {
+      "epoch": 4.236453201970443,
+      "grad_norm": 0.7681763768196106,
+      "learning_rate": 0.0002,
+      "loss": 0.5542,
+      "step": 3870
+    },
+    {
+      "epoch": 4.2474001094690745,
+      "grad_norm": 0.9840841293334961,
+      "learning_rate": 0.0002,
+      "loss": 0.5971,
+      "step": 3880
+    },
+    {
+      "epoch": 4.258347016967707,
+      "grad_norm": 1.0290725231170654,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 3890
+    },
+    {
+      "epoch": 4.269293924466338,
+      "grad_norm": 0.8059597611427307,
+      "learning_rate": 0.0002,
+      "loss": 0.4288,
+      "step": 3900
+    },
+    {
+      "epoch": 4.28024083196497,
+      "grad_norm": 0.9847467541694641,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3910
+    },
+    {
+      "epoch": 4.291187739463601,
+      "grad_norm": 1.344044804573059,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 3920
+    },
+    {
+      "epoch": 4.302134646962233,
+      "grad_norm": 0.9174224138259888,
+      "learning_rate": 0.0002,
+      "loss": 0.4966,
+      "step": 3930
+    },
+    {
+      "epoch": 4.313081554460865,
+      "grad_norm": 1.1199711561203003,
+      "learning_rate": 0.0002,
+      "loss": 0.4944,
+      "step": 3940
+    },
+    {
+      "epoch": 4.324028461959497,
+      "grad_norm": 1.0120296478271484,
+      "learning_rate": 0.0002,
+      "loss": 0.4641,
+      "step": 3950
+    },
+    {
+      "epoch": 4.334975369458128,
+      "grad_norm": 1.091811180114746,
+      "learning_rate": 0.0002,
+      "loss": 0.4723,
+      "step": 3960
+    },
+    {
+      "epoch": 4.34592227695676,
+      "grad_norm": 1.0332133769989014,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 3970
+    },
+    {
+      "epoch": 4.356869184455391,
+      "grad_norm": 1.0785295963287354,
+      "learning_rate": 0.0002,
+      "loss": 0.4646,
+      "step": 3980
+    },
+    {
+      "epoch": 4.3678160919540225,
+      "grad_norm": 1.0506969690322876,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 3990
+    },
+    {
+      "epoch": 4.378762999452655,
+      "grad_norm": 1.047560691833496,
+      "learning_rate": 0.0002,
+      "loss": 0.4776,
+      "step": 4000
+    },
+    {
+      "epoch": 4.389709906951286,
+      "grad_norm": 0.9348800778388977,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 4010
+    },
+    {
+      "epoch": 4.400656814449918,
+      "grad_norm": 1.1563059091567993,
+      "learning_rate": 0.0002,
+      "loss": 0.5333,
+      "step": 4020
+    },
+    {
+      "epoch": 4.411603721948549,
+      "grad_norm": 1.001470923423767,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 4030
+    },
+    {
+      "epoch": 4.422550629447181,
+      "grad_norm": 1.309012532234192,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 4040
+    },
+    {
+      "epoch": 4.433497536945813,
+      "grad_norm": 0.7338925004005432,
+      "learning_rate": 0.0002,
+      "loss": 0.5078,
+      "step": 4050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 1.0398834943771362,
+      "learning_rate": 0.0002,
+      "loss": 0.4632,
+      "step": 4060
+    },
+    {
+      "epoch": 4.455391351943076,
+      "grad_norm": 0.9728689193725586,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4070
+    },
+    {
+      "epoch": 4.466338259441708,
+      "grad_norm": 1.247475028038025,
+      "learning_rate": 0.0002,
+      "loss": 0.4778,
+      "step": 4080
+    },
+    {
+      "epoch": 4.477285166940339,
+      "grad_norm": 1.1084578037261963,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 4090
+    },
+    {
+      "epoch": 4.4882320744389705,
+      "grad_norm": 1.1619318723678589,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 4100
+    },
+    {
+      "epoch": 4.499178981937603,
+      "grad_norm": 1.3456498384475708,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4110
+    },
+    {
+      "epoch": 4.510125889436234,
+      "grad_norm": 0.9372991323471069,
+      "learning_rate": 0.0002,
+      "loss": 0.4876,
+      "step": 4120
+    },
+    {
+      "epoch": 4.521072796934866,
+      "grad_norm": 1.0071815252304077,
+      "learning_rate": 0.0002,
+      "loss": 0.5456,
+      "step": 4130
+    },
+    {
+      "epoch": 4.532019704433497,
+      "grad_norm": 1.190344214439392,
+      "learning_rate": 0.0002,
+      "loss": 0.5589,
+      "step": 4140
+    },
+    {
+      "epoch": 4.542966611932129,
+      "grad_norm": 0.9480887055397034,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 4150
+    },
+    {
+      "epoch": 4.553913519430761,
+      "grad_norm": 1.0252189636230469,
+      "learning_rate": 0.0002,
+      "loss": 0.5229,
+      "step": 4160
+    },
+    {
+      "epoch": 4.564860426929393,
+      "grad_norm": 0.7142013311386108,
+      "learning_rate": 0.0002,
+      "loss": 0.5253,
+      "step": 4170
+    },
+    {
+      "epoch": 4.575807334428024,
+      "grad_norm": 0.8937426805496216,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 4180
+    },
+    {
+      "epoch": 4.586754241926656,
+      "grad_norm": 0.8885005116462708,
+      "learning_rate": 0.0002,
+      "loss": 0.4773,
+      "step": 4190
+    },
+    {
+      "epoch": 4.597701149425287,
+      "grad_norm": 1.337663173675537,
+      "learning_rate": 0.0002,
+      "loss": 0.4858,
+      "step": 4200
+    },
+    {
+      "epoch": 4.6086480569239185,
+      "grad_norm": 1.0475375652313232,
+      "learning_rate": 0.0002,
+      "loss": 0.5247,
+      "step": 4210
+    },
+    {
+      "epoch": 4.619594964422551,
+      "grad_norm": 1.0081088542938232,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 4220
+    },
+    {
+      "epoch": 4.630541871921182,
+      "grad_norm": 0.7527595162391663,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 4230
+    },
+    {
+      "epoch": 4.641488779419814,
+      "grad_norm": 1.55559241771698,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4240
+    },
+    {
+      "epoch": 4.652435686918445,
+      "grad_norm": 0.7967379689216614,
+      "learning_rate": 0.0002,
+      "loss": 0.5468,
+      "step": 4250
+    },
+    {
+      "epoch": 4.663382594417077,
+      "grad_norm": 0.898368775844574,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 4260
+    },
+    {
+      "epoch": 4.674329501915709,
+      "grad_norm": 1.1940776109695435,
+      "learning_rate": 0.0002,
+      "loss": 0.4706,
+      "step": 4270
+    },
+    {
+      "epoch": 4.685276409414341,
+      "grad_norm": 1.1817092895507812,
+      "learning_rate": 0.0002,
+      "loss": 0.5121,
+      "step": 4280
+    },
+    {
+      "epoch": 4.696223316912972,
+      "grad_norm": 0.9041520357131958,
+      "learning_rate": 0.0002,
+      "loss": 0.5758,
+      "step": 4290
+    },
+    {
+      "epoch": 4.707170224411604,
+      "grad_norm": 1.1280102729797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5851,
+      "step": 4300
+    },
+    {
+      "epoch": 4.718117131910235,
+      "grad_norm": 1.357689619064331,
+      "learning_rate": 0.0002,
+      "loss": 0.4891,
+      "step": 4310
+    },
+    {
+      "epoch": 4.7290640394088665,
+      "grad_norm": 1.056633472442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4704,
+      "step": 4320
+    },
+    {
+      "epoch": 4.740010946907499,
+      "grad_norm": 1.6520427465438843,
+      "learning_rate": 0.0002,
+      "loss": 0.5488,
+      "step": 4330
+    },
+    {
+      "epoch": 4.75095785440613,
+      "grad_norm": 1.153200626373291,
+      "learning_rate": 0.0002,
+      "loss": 0.5131,
+      "step": 4340
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.9346241354942322,
+      "learning_rate": 0.0002,
+      "loss": 0.539,
+      "step": 4350
+    },
+    {
+      "epoch": 4.772851669403393,
+      "grad_norm": 0.8628455996513367,
+      "learning_rate": 0.0002,
+      "loss": 0.4941,
+      "step": 4360
+    },
+    {
+      "epoch": 4.783798576902025,
+      "grad_norm": 1.3843916654586792,
+      "learning_rate": 0.0002,
+      "loss": 0.5167,
+      "step": 4370
+    },
+    {
+      "epoch": 4.794745484400657,
+      "grad_norm": 1.035574197769165,
+      "learning_rate": 0.0002,
+      "loss": 0.4683,
+      "step": 4380
+    },
+    {
+      "epoch": 4.805692391899289,
+      "grad_norm": 1.1868361234664917,
+      "learning_rate": 0.0002,
+      "loss": 0.5162,
+      "step": 4390
+    },
+    {
+      "epoch": 4.81663929939792,
+      "grad_norm": 1.1307647228240967,
+      "learning_rate": 0.0002,
+      "loss": 0.534,
+      "step": 4400
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 0.9787724614143372,
+      "learning_rate": 0.0002,
+      "loss": 0.5567,
+      "step": 4410
+    },
+    {
+      "epoch": 4.838533114395183,
+      "grad_norm": 1.0473824739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.5185,
+      "step": 4420
+    },
+    {
+      "epoch": 4.8494800218938146,
+      "grad_norm": 1.069069504737854,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4430
+    },
+    {
+      "epoch": 4.860426929392447,
+      "grad_norm": 1.4305680990219116,
+      "learning_rate": 0.0002,
+      "loss": 0.5267,
+      "step": 4440
+    },
+    {
+      "epoch": 4.871373836891078,
+      "grad_norm": 1.3679203987121582,
+      "learning_rate": 0.0002,
+      "loss": 0.5947,
+      "step": 4450
+    },
+    {
+      "epoch": 4.88232074438971,
+      "grad_norm": 0.8997844457626343,
+      "learning_rate": 0.0002,
+      "loss": 0.5135,
+      "step": 4460
+    },
+    {
+      "epoch": 4.893267651888341,
+      "grad_norm": 1.2758110761642456,
+      "learning_rate": 0.0002,
+      "loss": 0.5312,
+      "step": 4470
+    },
+    {
+      "epoch": 4.904214559386973,
+      "grad_norm": 0.8819465637207031,
+      "learning_rate": 0.0002,
+      "loss": 0.4914,
+      "step": 4480
+    },
+    {
+      "epoch": 4.915161466885605,
+      "grad_norm": 1.08329439163208,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 4490
+    },
+    {
+      "epoch": 4.926108374384237,
+      "grad_norm": 1.083461046218872,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 4500
+    },
+    {
+      "epoch": 4.937055281882868,
+      "grad_norm": 1.2387723922729492,
+      "learning_rate": 0.0002,
+      "loss": 0.5433,
+      "step": 4510
+    },
+    {
+      "epoch": 4.9480021893815,
+      "grad_norm": 0.8262293934822083,
+      "learning_rate": 0.0002,
+      "loss": 0.5624,
+      "step": 4520
+    },
+    {
+      "epoch": 4.958949096880131,
+      "grad_norm": 1.2325191497802734,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 4530
+    },
+    {
+      "epoch": 4.9698960043787626,
+      "grad_norm": 1.024614930152893,
+      "learning_rate": 0.0002,
+      "loss": 0.5452,
+      "step": 4540
+    },
+    {
+      "epoch": 4.980842911877395,
+      "grad_norm": 1.3007521629333496,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 4550
+    },
+    {
+      "epoch": 4.991789819376026,
+      "grad_norm": 0.9823828339576721,
+      "learning_rate": 0.0002,
+      "loss": 0.4943,
+      "step": 4560
+    },
+    {
+      "epoch": 4.999452654625069,
+      "eval_loss": 1.3920727968215942,
+      "eval_runtime": 46.0764,
+      "eval_samples_per_second": 9.463,
+      "eval_steps_per_second": 1.194,
+      "step": 4567
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 2.346039081566208e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-4567/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..82c133a73dac49fd8b22dae8cd0105dd3e834f26
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:813c10c32a1db7b0343c09b57fe6ff198f4d9c5e705f4c0fa06a932d007912ae
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2fe267b14fe3a5b73d54b2a70f216c38551da26b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b546deea7f03b6244165ccd337e00769d9bf8ea1b1e7828006bbffc59607b9a
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ab79c8b7dfce179b8a9c1ac2327a60afed32b6eb
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b15c5d80fbe79a987a0ec29b50d9d5373326b9de9985c638f7efcbd9ed4177ec
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8c4e019fb2c3b15fc765ed001d4d756a29a9f82d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9d21638e7e3836177ef804d71bacd5b91faa871fce0cf0ea4b7895c8599e125
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..996adc3e8b127f52fea698ba5b1a4699f222fd50
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/trainer_state.json
@@ -0,0 +1,3917 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 5481,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    },
+    {
+      "epoch": 3.0103995621237,
+      "grad_norm": 0.6384502053260803,
+      "learning_rate": 0.0002,
+      "loss": 0.6443,
+      "step": 2750
+    },
+    {
+      "epoch": 3.0213464696223316,
+      "grad_norm": 0.9928722381591797,
+      "learning_rate": 0.0002,
+      "loss": 0.7123,
+      "step": 2760
+    },
+    {
+      "epoch": 3.0322933771209635,
+      "grad_norm": 0.7813051342964172,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 2770
+    },
+    {
+      "epoch": 3.043240284619595,
+      "grad_norm": 1.0202556848526,
+      "learning_rate": 0.0002,
+      "loss": 0.6042,
+      "step": 2780
+    },
+    {
+      "epoch": 3.0541871921182264,
+      "grad_norm": 0.7581062316894531,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 2790
+    },
+    {
+      "epoch": 3.0651340996168583,
+      "grad_norm": 0.6252710223197937,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 2800
+    },
+    {
+      "epoch": 3.07608100711549,
+      "grad_norm": 0.7738662958145142,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 2810
+    },
+    {
+      "epoch": 3.0870279146141213,
+      "grad_norm": 0.7381885051727295,
+      "learning_rate": 0.0002,
+      "loss": 0.627,
+      "step": 2820
+    },
+    {
+      "epoch": 3.097974822112753,
+      "grad_norm": 0.9197564721107483,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 2830
+    },
+    {
+      "epoch": 3.1089217296113847,
+      "grad_norm": 1.000976800918579,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 2840
+    },
+    {
+      "epoch": 3.1198686371100166,
+      "grad_norm": 0.7559131383895874,
+      "learning_rate": 0.0002,
+      "loss": 0.6631,
+      "step": 2850
+    },
+    {
+      "epoch": 3.130815544608648,
+      "grad_norm": 0.7213780879974365,
+      "learning_rate": 0.0002,
+      "loss": 0.6252,
+      "step": 2860
+    },
+    {
+      "epoch": 3.1417624521072796,
+      "grad_norm": 0.945939838886261,
+      "learning_rate": 0.0002,
+      "loss": 0.6501,
+      "step": 2870
+    },
+    {
+      "epoch": 3.1527093596059115,
+      "grad_norm": 0.7277454137802124,
+      "learning_rate": 0.0002,
+      "loss": 0.6129,
+      "step": 2880
+    },
+    {
+      "epoch": 3.163656267104543,
+      "grad_norm": 0.762026846408844,
+      "learning_rate": 0.0002,
+      "loss": 0.6423,
+      "step": 2890
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6471221446990967,
+      "learning_rate": 0.0002,
+      "loss": 0.5332,
+      "step": 2900
+    },
+    {
+      "epoch": 3.1855500821018063,
+      "grad_norm": 0.6018978357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 2910
+    },
+    {
+      "epoch": 3.196496989600438,
+      "grad_norm": 0.8607320785522461,
+      "learning_rate": 0.0002,
+      "loss": 0.7274,
+      "step": 2920
+    },
+    {
+      "epoch": 3.2074438970990693,
+      "grad_norm": 0.8854126334190369,
+      "learning_rate": 0.0002,
+      "loss": 0.6139,
+      "step": 2930
+    },
+    {
+      "epoch": 3.218390804597701,
+      "grad_norm": 0.6620870232582092,
+      "learning_rate": 0.0002,
+      "loss": 0.6485,
+      "step": 2940
+    },
+    {
+      "epoch": 3.2293377120963327,
+      "grad_norm": 0.7377511858940125,
+      "learning_rate": 0.0002,
+      "loss": 0.6969,
+      "step": 2950
+    },
+    {
+      "epoch": 3.2402846195949646,
+      "grad_norm": 0.7803301811218262,
+      "learning_rate": 0.0002,
+      "loss": 0.6798,
+      "step": 2960
+    },
+    {
+      "epoch": 3.251231527093596,
+      "grad_norm": 0.834061861038208,
+      "learning_rate": 0.0002,
+      "loss": 0.6697,
+      "step": 2970
+    },
+    {
+      "epoch": 3.2621784345922276,
+      "grad_norm": 0.8496041893959045,
+      "learning_rate": 0.0002,
+      "loss": 0.6894,
+      "step": 2980
+    },
+    {
+      "epoch": 3.2731253420908595,
+      "grad_norm": 0.7967984676361084,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 2990
+    },
+    {
+      "epoch": 3.284072249589491,
+      "grad_norm": 1.0207016468048096,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 3000
+    },
+    {
+      "epoch": 3.2950191570881224,
+      "grad_norm": 0.7855771780014038,
+      "learning_rate": 0.0002,
+      "loss": 0.6586,
+      "step": 3010
+    },
+    {
+      "epoch": 3.3059660645867543,
+      "grad_norm": 0.9427546858787537,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 3020
+    },
+    {
+      "epoch": 3.316912972085386,
+      "grad_norm": 0.823542594909668,
+      "learning_rate": 0.0002,
+      "loss": 0.6277,
+      "step": 3030
+    },
+    {
+      "epoch": 3.3278598795840173,
+      "grad_norm": 0.9826635122299194,
+      "learning_rate": 0.0002,
+      "loss": 0.7109,
+      "step": 3040
+    },
+    {
+      "epoch": 3.338806787082649,
+      "grad_norm": 0.7259827852249146,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 3050
+    },
+    {
+      "epoch": 3.3497536945812807,
+      "grad_norm": 0.7774739861488342,
+      "learning_rate": 0.0002,
+      "loss": 0.653,
+      "step": 3060
+    },
+    {
+      "epoch": 3.3607006020799126,
+      "grad_norm": 0.7394293546676636,
+      "learning_rate": 0.0002,
+      "loss": 0.7529,
+      "step": 3070
+    },
+    {
+      "epoch": 3.371647509578544,
+      "grad_norm": 0.9017578959465027,
+      "learning_rate": 0.0002,
+      "loss": 0.5987,
+      "step": 3080
+    },
+    {
+      "epoch": 3.3825944170771756,
+      "grad_norm": 0.7451054453849792,
+      "learning_rate": 0.0002,
+      "loss": 0.6953,
+      "step": 3090
+    },
+    {
+      "epoch": 3.3935413245758075,
+      "grad_norm": 0.7321506142616272,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 3100
+    },
+    {
+      "epoch": 3.404488232074439,
+      "grad_norm": 0.6721828579902649,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3110
+    },
+    {
+      "epoch": 3.4154351395730704,
+      "grad_norm": 0.774022102355957,
+      "learning_rate": 0.0002,
+      "loss": 0.6559,
+      "step": 3120
+    },
+    {
+      "epoch": 3.4263820470717024,
+      "grad_norm": 0.9143537282943726,
+      "learning_rate": 0.0002,
+      "loss": 0.7449,
+      "step": 3130
+    },
+    {
+      "epoch": 3.437328954570334,
+      "grad_norm": 1.226087212562561,
+      "learning_rate": 0.0002,
+      "loss": 0.6899,
+      "step": 3140
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7545496225357056,
+      "learning_rate": 0.0002,
+      "loss": 0.6719,
+      "step": 3150
+    },
+    {
+      "epoch": 3.4592227695675972,
+      "grad_norm": 0.6515635848045349,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 3160
+    },
+    {
+      "epoch": 3.4701696770662287,
+      "grad_norm": 0.9297090172767639,
+      "learning_rate": 0.0002,
+      "loss": 0.6926,
+      "step": 3170
+    },
+    {
+      "epoch": 3.4811165845648606,
+      "grad_norm": 1.0130730867385864,
+      "learning_rate": 0.0002,
+      "loss": 0.6071,
+      "step": 3180
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.7654589414596558,
+      "learning_rate": 0.0002,
+      "loss": 0.5959,
+      "step": 3190
+    },
+    {
+      "epoch": 3.5030103995621236,
+      "grad_norm": 0.9954977631568909,
+      "learning_rate": 0.0002,
+      "loss": 0.7401,
+      "step": 3200
+    },
+    {
+      "epoch": 3.5139573070607555,
+      "grad_norm": 0.6027487516403198,
+      "learning_rate": 0.0002,
+      "loss": 0.6661,
+      "step": 3210
+    },
+    {
+      "epoch": 3.524904214559387,
+      "grad_norm": 0.741770327091217,
+      "learning_rate": 0.0002,
+      "loss": 0.6963,
+      "step": 3220
+    },
+    {
+      "epoch": 3.535851122058019,
+      "grad_norm": 1.0534909963607788,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 3230
+    },
+    {
+      "epoch": 3.5467980295566504,
+      "grad_norm": 0.937772274017334,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 3240
+    },
+    {
+      "epoch": 3.557744937055282,
+      "grad_norm": 0.8504213690757751,
+      "learning_rate": 0.0002,
+      "loss": 0.6681,
+      "step": 3250
+    },
+    {
+      "epoch": 3.5686918445539133,
+      "grad_norm": 0.7755007147789001,
+      "learning_rate": 0.0002,
+      "loss": 0.6436,
+      "step": 3260
+    },
+    {
+      "epoch": 3.5796387520525452,
+      "grad_norm": 1.0193358659744263,
+      "learning_rate": 0.0002,
+      "loss": 0.6213,
+      "step": 3270
+    },
+    {
+      "epoch": 3.5905856595511767,
+      "grad_norm": 0.8440536856651306,
+      "learning_rate": 0.0002,
+      "loss": 0.671,
+      "step": 3280
+    },
+    {
+      "epoch": 3.6015325670498086,
+      "grad_norm": 0.6195939183235168,
+      "learning_rate": 0.0002,
+      "loss": 0.6859,
+      "step": 3290
+    },
+    {
+      "epoch": 3.61247947454844,
+      "grad_norm": 0.8608590960502625,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3300
+    },
+    {
+      "epoch": 3.6234263820470716,
+      "grad_norm": 0.6772327423095703,
+      "learning_rate": 0.0002,
+      "loss": 0.7301,
+      "step": 3310
+    },
+    {
+      "epoch": 3.6343732895457035,
+      "grad_norm": 0.8031839728355408,
+      "learning_rate": 0.0002,
+      "loss": 0.6298,
+      "step": 3320
+    },
+    {
+      "epoch": 3.645320197044335,
+      "grad_norm": 0.6080502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.7041,
+      "step": 3330
+    },
+    {
+      "epoch": 3.656267104542967,
+      "grad_norm": 0.8007240891456604,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 3340
+    },
+    {
+      "epoch": 3.6672140120415984,
+      "grad_norm": 0.8060704469680786,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3350
+    },
+    {
+      "epoch": 3.67816091954023,
+      "grad_norm": 0.7547586560249329,
+      "learning_rate": 0.0002,
+      "loss": 0.6304,
+      "step": 3360
+    },
+    {
+      "epoch": 3.6891078270388613,
+      "grad_norm": 0.686851978302002,
+      "learning_rate": 0.0002,
+      "loss": 0.7066,
+      "step": 3370
+    },
+    {
+      "epoch": 3.7000547345374932,
+      "grad_norm": 0.9429075717926025,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 3380
+    },
+    {
+      "epoch": 3.7110016420361247,
+      "grad_norm": 0.7283591032028198,
+      "learning_rate": 0.0002,
+      "loss": 0.6673,
+      "step": 3390
+    },
+    {
+      "epoch": 3.7219485495347566,
+      "grad_norm": 0.8323085904121399,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 3400
+    },
+    {
+      "epoch": 3.732895457033388,
+      "grad_norm": 0.8529590964317322,
+      "learning_rate": 0.0002,
+      "loss": 0.7779,
+      "step": 3410
+    },
+    {
+      "epoch": 3.7438423645320196,
+      "grad_norm": 0.731752872467041,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3420
+    },
+    {
+      "epoch": 3.7547892720306515,
+      "grad_norm": 0.8572278618812561,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3430
+    },
+    {
+      "epoch": 3.765736179529283,
+      "grad_norm": 0.7408691048622131,
+      "learning_rate": 0.0002,
+      "loss": 0.6215,
+      "step": 3440
+    },
+    {
+      "epoch": 3.776683087027915,
+      "grad_norm": 0.7470445036888123,
+      "learning_rate": 0.0002,
+      "loss": 0.622,
+      "step": 3450
+    },
+    {
+      "epoch": 3.7876299945265464,
+      "grad_norm": 0.6806244254112244,
+      "learning_rate": 0.0002,
+      "loss": 0.7241,
+      "step": 3460
+    },
+    {
+      "epoch": 3.798576902025178,
+      "grad_norm": 0.9129069447517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7739,
+      "step": 3470
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.8717501759529114,
+      "learning_rate": 0.0002,
+      "loss": 0.6826,
+      "step": 3480
+    },
+    {
+      "epoch": 3.8204707170224412,
+      "grad_norm": 0.6761979460716248,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 3490
+    },
+    {
+      "epoch": 3.8314176245210727,
+      "grad_norm": 1.0054380893707275,
+      "learning_rate": 0.0002,
+      "loss": 0.7601,
+      "step": 3500
+    },
+    {
+      "epoch": 3.8423645320197046,
+      "grad_norm": 1.1224009990692139,
+      "learning_rate": 0.0002,
+      "loss": 0.623,
+      "step": 3510
+    },
+    {
+      "epoch": 3.853311439518336,
+      "grad_norm": 0.8997692465782166,
+      "learning_rate": 0.0002,
+      "loss": 0.6918,
+      "step": 3520
+    },
+    {
+      "epoch": 3.8642583470169676,
+      "grad_norm": 1.0086902379989624,
+      "learning_rate": 0.0002,
+      "loss": 0.6357,
+      "step": 3530
+    },
+    {
+      "epoch": 3.8752052545155995,
+      "grad_norm": 0.772739589214325,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 3540
+    },
+    {
+      "epoch": 3.886152162014231,
+      "grad_norm": 1.211774230003357,
+      "learning_rate": 0.0002,
+      "loss": 0.7423,
+      "step": 3550
+    },
+    {
+      "epoch": 3.897099069512863,
+      "grad_norm": 0.9572356939315796,
+      "learning_rate": 0.0002,
+      "loss": 0.7321,
+      "step": 3560
+    },
+    {
+      "epoch": 3.9080459770114944,
+      "grad_norm": 0.7887842655181885,
+      "learning_rate": 0.0002,
+      "loss": 0.6836,
+      "step": 3570
+    },
+    {
+      "epoch": 3.918992884510126,
+      "grad_norm": 0.7308389544487,
+      "learning_rate": 0.0002,
+      "loss": 0.7576,
+      "step": 3580
+    },
+    {
+      "epoch": 3.9299397920087573,
+      "grad_norm": 1.0182650089263916,
+      "learning_rate": 0.0002,
+      "loss": 0.6001,
+      "step": 3590
+    },
+    {
+      "epoch": 3.9408866995073892,
+      "grad_norm": 0.8000147342681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 3600
+    },
+    {
+      "epoch": 3.9518336070060207,
+      "grad_norm": 0.7385728359222412,
+      "learning_rate": 0.0002,
+      "loss": 0.6244,
+      "step": 3610
+    },
+    {
+      "epoch": 3.9627805145046526,
+      "grad_norm": 0.9233261942863464,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 3620
+    },
+    {
+      "epoch": 3.973727422003284,
+      "grad_norm": 0.8486751914024353,
+      "learning_rate": 0.0002,
+      "loss": 0.6508,
+      "step": 3630
+    },
+    {
+      "epoch": 3.9846743295019156,
+      "grad_norm": 0.7593663334846497,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3640
+    },
+    {
+      "epoch": 3.9956212370005475,
+      "grad_norm": 0.7885415554046631,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 3650
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.250312328338623,
+      "eval_runtime": 46.0842,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.193,
+      "step": 3654
+    },
+    {
+      "epoch": 4.006568144499179,
+      "grad_norm": 0.6591703295707703,
+      "learning_rate": 0.0002,
+      "loss": 0.5547,
+      "step": 3660
+    },
+    {
+      "epoch": 4.017515051997811,
+      "grad_norm": 1.36927330493927,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 3670
+    },
+    {
+      "epoch": 4.028461959496442,
+      "grad_norm": 0.8106328845024109,
+      "learning_rate": 0.0002,
+      "loss": 0.4466,
+      "step": 3680
+    },
+    {
+      "epoch": 4.039408866995074,
+      "grad_norm": 0.7592712044715881,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 3690
+    },
+    {
+      "epoch": 4.050355774493705,
+      "grad_norm": 0.9518909454345703,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3700
+    },
+    {
+      "epoch": 4.061302681992337,
+      "grad_norm": 0.7805967330932617,
+      "learning_rate": 0.0002,
+      "loss": 0.4638,
+      "step": 3710
+    },
+    {
+      "epoch": 4.072249589490969,
+      "grad_norm": 1.3146334886550903,
+      "learning_rate": 0.0002,
+      "loss": 0.4556,
+      "step": 3720
+    },
+    {
+      "epoch": 4.083196496989601,
+      "grad_norm": 1.1611138582229614,
+      "learning_rate": 0.0002,
+      "loss": 0.5635,
+      "step": 3730
+    },
+    {
+      "epoch": 4.094143404488232,
+      "grad_norm": 0.8173232078552246,
+      "learning_rate": 0.0002,
+      "loss": 0.3845,
+      "step": 3740
+    },
+    {
+      "epoch": 4.105090311986864,
+      "grad_norm": 0.7848323583602905,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 3750
+    },
+    {
+      "epoch": 4.116037219485495,
+      "grad_norm": 1.3183201551437378,
+      "learning_rate": 0.0002,
+      "loss": 0.4519,
+      "step": 3760
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 1.1936529874801636,
+      "learning_rate": 0.0002,
+      "loss": 0.5083,
+      "step": 3770
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1078993082046509,
+      "learning_rate": 0.0002,
+      "loss": 0.5208,
+      "step": 3780
+    },
+    {
+      "epoch": 4.14887794198139,
+      "grad_norm": 1.107743263244629,
+      "learning_rate": 0.0002,
+      "loss": 0.5928,
+      "step": 3790
+    },
+    {
+      "epoch": 4.159824849480022,
+      "grad_norm": 0.7801875472068787,
+      "learning_rate": 0.0002,
+      "loss": 0.5112,
+      "step": 3800
+    },
+    {
+      "epoch": 4.170771756978653,
+      "grad_norm": 1.1328117847442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4896,
+      "step": 3810
+    },
+    {
+      "epoch": 4.181718664477285,
+      "grad_norm": 1.4232193231582642,
+      "learning_rate": 0.0002,
+      "loss": 0.5645,
+      "step": 3820
+    },
+    {
+      "epoch": 4.192665571975917,
+      "grad_norm": 1.557416558265686,
+      "learning_rate": 0.0002,
+      "loss": 0.5049,
+      "step": 3830
+    },
+    {
+      "epoch": 4.203612479474549,
+      "grad_norm": 1.042923092842102,
+      "learning_rate": 0.0002,
+      "loss": 0.4863,
+      "step": 3840
+    },
+    {
+      "epoch": 4.21455938697318,
+      "grad_norm": 1.1801949739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 3850
+    },
+    {
+      "epoch": 4.225506294471812,
+      "grad_norm": 0.9273753762245178,
+      "learning_rate": 0.0002,
+      "loss": 0.5063,
+      "step": 3860
+    },
+    {
+      "epoch": 4.236453201970443,
+      "grad_norm": 0.7681763768196106,
+      "learning_rate": 0.0002,
+      "loss": 0.5542,
+      "step": 3870
+    },
+    {
+      "epoch": 4.2474001094690745,
+      "grad_norm": 0.9840841293334961,
+      "learning_rate": 0.0002,
+      "loss": 0.5971,
+      "step": 3880
+    },
+    {
+      "epoch": 4.258347016967707,
+      "grad_norm": 1.0290725231170654,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 3890
+    },
+    {
+      "epoch": 4.269293924466338,
+      "grad_norm": 0.8059597611427307,
+      "learning_rate": 0.0002,
+      "loss": 0.4288,
+      "step": 3900
+    },
+    {
+      "epoch": 4.28024083196497,
+      "grad_norm": 0.9847467541694641,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3910
+    },
+    {
+      "epoch": 4.291187739463601,
+      "grad_norm": 1.344044804573059,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 3920
+    },
+    {
+      "epoch": 4.302134646962233,
+      "grad_norm": 0.9174224138259888,
+      "learning_rate": 0.0002,
+      "loss": 0.4966,
+      "step": 3930
+    },
+    {
+      "epoch": 4.313081554460865,
+      "grad_norm": 1.1199711561203003,
+      "learning_rate": 0.0002,
+      "loss": 0.4944,
+      "step": 3940
+    },
+    {
+      "epoch": 4.324028461959497,
+      "grad_norm": 1.0120296478271484,
+      "learning_rate": 0.0002,
+      "loss": 0.4641,
+      "step": 3950
+    },
+    {
+      "epoch": 4.334975369458128,
+      "grad_norm": 1.091811180114746,
+      "learning_rate": 0.0002,
+      "loss": 0.4723,
+      "step": 3960
+    },
+    {
+      "epoch": 4.34592227695676,
+      "grad_norm": 1.0332133769989014,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 3970
+    },
+    {
+      "epoch": 4.356869184455391,
+      "grad_norm": 1.0785295963287354,
+      "learning_rate": 0.0002,
+      "loss": 0.4646,
+      "step": 3980
+    },
+    {
+      "epoch": 4.3678160919540225,
+      "grad_norm": 1.0506969690322876,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 3990
+    },
+    {
+      "epoch": 4.378762999452655,
+      "grad_norm": 1.047560691833496,
+      "learning_rate": 0.0002,
+      "loss": 0.4776,
+      "step": 4000
+    },
+    {
+      "epoch": 4.389709906951286,
+      "grad_norm": 0.9348800778388977,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 4010
+    },
+    {
+      "epoch": 4.400656814449918,
+      "grad_norm": 1.1563059091567993,
+      "learning_rate": 0.0002,
+      "loss": 0.5333,
+      "step": 4020
+    },
+    {
+      "epoch": 4.411603721948549,
+      "grad_norm": 1.001470923423767,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 4030
+    },
+    {
+      "epoch": 4.422550629447181,
+      "grad_norm": 1.309012532234192,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 4040
+    },
+    {
+      "epoch": 4.433497536945813,
+      "grad_norm": 0.7338925004005432,
+      "learning_rate": 0.0002,
+      "loss": 0.5078,
+      "step": 4050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 1.0398834943771362,
+      "learning_rate": 0.0002,
+      "loss": 0.4632,
+      "step": 4060
+    },
+    {
+      "epoch": 4.455391351943076,
+      "grad_norm": 0.9728689193725586,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4070
+    },
+    {
+      "epoch": 4.466338259441708,
+      "grad_norm": 1.247475028038025,
+      "learning_rate": 0.0002,
+      "loss": 0.4778,
+      "step": 4080
+    },
+    {
+      "epoch": 4.477285166940339,
+      "grad_norm": 1.1084578037261963,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 4090
+    },
+    {
+      "epoch": 4.4882320744389705,
+      "grad_norm": 1.1619318723678589,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 4100
+    },
+    {
+      "epoch": 4.499178981937603,
+      "grad_norm": 1.3456498384475708,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4110
+    },
+    {
+      "epoch": 4.510125889436234,
+      "grad_norm": 0.9372991323471069,
+      "learning_rate": 0.0002,
+      "loss": 0.4876,
+      "step": 4120
+    },
+    {
+      "epoch": 4.521072796934866,
+      "grad_norm": 1.0071815252304077,
+      "learning_rate": 0.0002,
+      "loss": 0.5456,
+      "step": 4130
+    },
+    {
+      "epoch": 4.532019704433497,
+      "grad_norm": 1.190344214439392,
+      "learning_rate": 0.0002,
+      "loss": 0.5589,
+      "step": 4140
+    },
+    {
+      "epoch": 4.542966611932129,
+      "grad_norm": 0.9480887055397034,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 4150
+    },
+    {
+      "epoch": 4.553913519430761,
+      "grad_norm": 1.0252189636230469,
+      "learning_rate": 0.0002,
+      "loss": 0.5229,
+      "step": 4160
+    },
+    {
+      "epoch": 4.564860426929393,
+      "grad_norm": 0.7142013311386108,
+      "learning_rate": 0.0002,
+      "loss": 0.5253,
+      "step": 4170
+    },
+    {
+      "epoch": 4.575807334428024,
+      "grad_norm": 0.8937426805496216,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 4180
+    },
+    {
+      "epoch": 4.586754241926656,
+      "grad_norm": 0.8885005116462708,
+      "learning_rate": 0.0002,
+      "loss": 0.4773,
+      "step": 4190
+    },
+    {
+      "epoch": 4.597701149425287,
+      "grad_norm": 1.337663173675537,
+      "learning_rate": 0.0002,
+      "loss": 0.4858,
+      "step": 4200
+    },
+    {
+      "epoch": 4.6086480569239185,
+      "grad_norm": 1.0475375652313232,
+      "learning_rate": 0.0002,
+      "loss": 0.5247,
+      "step": 4210
+    },
+    {
+      "epoch": 4.619594964422551,
+      "grad_norm": 1.0081088542938232,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 4220
+    },
+    {
+      "epoch": 4.630541871921182,
+      "grad_norm": 0.7527595162391663,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 4230
+    },
+    {
+      "epoch": 4.641488779419814,
+      "grad_norm": 1.55559241771698,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4240
+    },
+    {
+      "epoch": 4.652435686918445,
+      "grad_norm": 0.7967379689216614,
+      "learning_rate": 0.0002,
+      "loss": 0.5468,
+      "step": 4250
+    },
+    {
+      "epoch": 4.663382594417077,
+      "grad_norm": 0.898368775844574,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 4260
+    },
+    {
+      "epoch": 4.674329501915709,
+      "grad_norm": 1.1940776109695435,
+      "learning_rate": 0.0002,
+      "loss": 0.4706,
+      "step": 4270
+    },
+    {
+      "epoch": 4.685276409414341,
+      "grad_norm": 1.1817092895507812,
+      "learning_rate": 0.0002,
+      "loss": 0.5121,
+      "step": 4280
+    },
+    {
+      "epoch": 4.696223316912972,
+      "grad_norm": 0.9041520357131958,
+      "learning_rate": 0.0002,
+      "loss": 0.5758,
+      "step": 4290
+    },
+    {
+      "epoch": 4.707170224411604,
+      "grad_norm": 1.1280102729797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5851,
+      "step": 4300
+    },
+    {
+      "epoch": 4.718117131910235,
+      "grad_norm": 1.357689619064331,
+      "learning_rate": 0.0002,
+      "loss": 0.4891,
+      "step": 4310
+    },
+    {
+      "epoch": 4.7290640394088665,
+      "grad_norm": 1.056633472442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4704,
+      "step": 4320
+    },
+    {
+      "epoch": 4.740010946907499,
+      "grad_norm": 1.6520427465438843,
+      "learning_rate": 0.0002,
+      "loss": 0.5488,
+      "step": 4330
+    },
+    {
+      "epoch": 4.75095785440613,
+      "grad_norm": 1.153200626373291,
+      "learning_rate": 0.0002,
+      "loss": 0.5131,
+      "step": 4340
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.9346241354942322,
+      "learning_rate": 0.0002,
+      "loss": 0.539,
+      "step": 4350
+    },
+    {
+      "epoch": 4.772851669403393,
+      "grad_norm": 0.8628455996513367,
+      "learning_rate": 0.0002,
+      "loss": 0.4941,
+      "step": 4360
+    },
+    {
+      "epoch": 4.783798576902025,
+      "grad_norm": 1.3843916654586792,
+      "learning_rate": 0.0002,
+      "loss": 0.5167,
+      "step": 4370
+    },
+    {
+      "epoch": 4.794745484400657,
+      "grad_norm": 1.035574197769165,
+      "learning_rate": 0.0002,
+      "loss": 0.4683,
+      "step": 4380
+    },
+    {
+      "epoch": 4.805692391899289,
+      "grad_norm": 1.1868361234664917,
+      "learning_rate": 0.0002,
+      "loss": 0.5162,
+      "step": 4390
+    },
+    {
+      "epoch": 4.81663929939792,
+      "grad_norm": 1.1307647228240967,
+      "learning_rate": 0.0002,
+      "loss": 0.534,
+      "step": 4400
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 0.9787724614143372,
+      "learning_rate": 0.0002,
+      "loss": 0.5567,
+      "step": 4410
+    },
+    {
+      "epoch": 4.838533114395183,
+      "grad_norm": 1.0473824739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.5185,
+      "step": 4420
+    },
+    {
+      "epoch": 4.8494800218938146,
+      "grad_norm": 1.069069504737854,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4430
+    },
+    {
+      "epoch": 4.860426929392447,
+      "grad_norm": 1.4305680990219116,
+      "learning_rate": 0.0002,
+      "loss": 0.5267,
+      "step": 4440
+    },
+    {
+      "epoch": 4.871373836891078,
+      "grad_norm": 1.3679203987121582,
+      "learning_rate": 0.0002,
+      "loss": 0.5947,
+      "step": 4450
+    },
+    {
+      "epoch": 4.88232074438971,
+      "grad_norm": 0.8997844457626343,
+      "learning_rate": 0.0002,
+      "loss": 0.5135,
+      "step": 4460
+    },
+    {
+      "epoch": 4.893267651888341,
+      "grad_norm": 1.2758110761642456,
+      "learning_rate": 0.0002,
+      "loss": 0.5312,
+      "step": 4470
+    },
+    {
+      "epoch": 4.904214559386973,
+      "grad_norm": 0.8819465637207031,
+      "learning_rate": 0.0002,
+      "loss": 0.4914,
+      "step": 4480
+    },
+    {
+      "epoch": 4.915161466885605,
+      "grad_norm": 1.08329439163208,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 4490
+    },
+    {
+      "epoch": 4.926108374384237,
+      "grad_norm": 1.083461046218872,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 4500
+    },
+    {
+      "epoch": 4.937055281882868,
+      "grad_norm": 1.2387723922729492,
+      "learning_rate": 0.0002,
+      "loss": 0.5433,
+      "step": 4510
+    },
+    {
+      "epoch": 4.9480021893815,
+      "grad_norm": 0.8262293934822083,
+      "learning_rate": 0.0002,
+      "loss": 0.5624,
+      "step": 4520
+    },
+    {
+      "epoch": 4.958949096880131,
+      "grad_norm": 1.2325191497802734,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 4530
+    },
+    {
+      "epoch": 4.9698960043787626,
+      "grad_norm": 1.024614930152893,
+      "learning_rate": 0.0002,
+      "loss": 0.5452,
+      "step": 4540
+    },
+    {
+      "epoch": 4.980842911877395,
+      "grad_norm": 1.3007521629333496,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 4550
+    },
+    {
+      "epoch": 4.991789819376026,
+      "grad_norm": 0.9823828339576721,
+      "learning_rate": 0.0002,
+      "loss": 0.4943,
+      "step": 4560
+    },
+    {
+      "epoch": 4.999452654625069,
+      "eval_loss": 1.3920727968215942,
+      "eval_runtime": 46.0764,
+      "eval_samples_per_second": 9.463,
+      "eval_steps_per_second": 1.194,
+      "step": 4567
+    },
+    {
+      "epoch": 5.002736726874658,
+      "grad_norm": 1.1478906869888306,
+      "learning_rate": 0.0002,
+      "loss": 0.545,
+      "step": 4570
+    },
+    {
+      "epoch": 5.013683634373289,
+      "grad_norm": 1.0533705949783325,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 4580
+    },
+    {
+      "epoch": 5.024630541871921,
+      "grad_norm": 1.268900752067566,
+      "learning_rate": 0.0002,
+      "loss": 0.3313,
+      "step": 4590
+    },
+    {
+      "epoch": 5.035577449370553,
+      "grad_norm": 1.222652554512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3482,
+      "step": 4600
+    },
+    {
+      "epoch": 5.046524356869185,
+      "grad_norm": 1.5093127489089966,
+      "learning_rate": 0.0002,
+      "loss": 0.3195,
+      "step": 4610
+    },
+    {
+      "epoch": 5.057471264367816,
+      "grad_norm": 1.2372499704360962,
+      "learning_rate": 0.0002,
+      "loss": 0.3569,
+      "step": 4620
+    },
+    {
+      "epoch": 5.068418171866448,
+      "grad_norm": 0.8422666192054749,
+      "learning_rate": 0.0002,
+      "loss": 0.3206,
+      "step": 4630
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 1.1451770067214966,
+      "learning_rate": 0.0002,
+      "loss": 0.3115,
+      "step": 4640
+    },
+    {
+      "epoch": 5.090311986863711,
+      "grad_norm": 1.2074557542800903,
+      "learning_rate": 0.0002,
+      "loss": 0.3305,
+      "step": 4650
+    },
+    {
+      "epoch": 5.101258894362343,
+      "grad_norm": 1.429150104522705,
+      "learning_rate": 0.0002,
+      "loss": 0.3012,
+      "step": 4660
+    },
+    {
+      "epoch": 5.112205801860974,
+      "grad_norm": 1.0353610515594482,
+      "learning_rate": 0.0002,
+      "loss": 0.3229,
+      "step": 4670
+    },
+    {
+      "epoch": 5.123152709359606,
+      "grad_norm": 1.2845979928970337,
+      "learning_rate": 0.0002,
+      "loss": 0.402,
+      "step": 4680
+    },
+    {
+      "epoch": 5.134099616858237,
+      "grad_norm": 1.3790186643600464,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 4690
+    },
+    {
+      "epoch": 5.145046524356869,
+      "grad_norm": 1.3182239532470703,
+      "learning_rate": 0.0002,
+      "loss": 0.2951,
+      "step": 4700
+    },
+    {
+      "epoch": 5.155993431855501,
+      "grad_norm": 1.5249626636505127,
+      "learning_rate": 0.0002,
+      "loss": 0.4074,
+      "step": 4710
+    },
+    {
+      "epoch": 5.166940339354133,
+      "grad_norm": 1.2492733001708984,
+      "learning_rate": 0.0002,
+      "loss": 0.3703,
+      "step": 4720
+    },
+    {
+      "epoch": 5.177887246852764,
+      "grad_norm": 1.4455480575561523,
+      "learning_rate": 0.0002,
+      "loss": 0.3411,
+      "step": 4730
+    },
+    {
+      "epoch": 5.188834154351396,
+      "grad_norm": 1.2191482782363892,
+      "learning_rate": 0.0002,
+      "loss": 0.3996,
+      "step": 4740
+    },
+    {
+      "epoch": 5.199781061850027,
+      "grad_norm": 1.4707951545715332,
+      "learning_rate": 0.0002,
+      "loss": 0.3785,
+      "step": 4750
+    },
+    {
+      "epoch": 5.210727969348659,
+      "grad_norm": 1.3473678827285767,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4760
+    },
+    {
+      "epoch": 5.221674876847291,
+      "grad_norm": 1.0479670763015747,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4770
+    },
+    {
+      "epoch": 5.232621784345922,
+      "grad_norm": 1.299096703529358,
+      "learning_rate": 0.0002,
+      "loss": 0.3976,
+      "step": 4780
+    },
+    {
+      "epoch": 5.243568691844554,
+      "grad_norm": 1.2820168733596802,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4790
+    },
+    {
+      "epoch": 5.254515599343185,
+      "grad_norm": 1.3818004131317139,
+      "learning_rate": 0.0002,
+      "loss": 0.3347,
+      "step": 4800
+    },
+    {
+      "epoch": 5.265462506841817,
+      "grad_norm": 1.2898736000061035,
+      "learning_rate": 0.0002,
+      "loss": 0.3761,
+      "step": 4810
+    },
+    {
+      "epoch": 5.276409414340449,
+      "grad_norm": 1.1761468648910522,
+      "learning_rate": 0.0002,
+      "loss": 0.3694,
+      "step": 4820
+    },
+    {
+      "epoch": 5.287356321839081,
+      "grad_norm": 1.7155952453613281,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4830
+    },
+    {
+      "epoch": 5.298303229337712,
+      "grad_norm": 0.9103642106056213,
+      "learning_rate": 0.0002,
+      "loss": 0.322,
+      "step": 4840
+    },
+    {
+      "epoch": 5.309250136836344,
+      "grad_norm": 1.013015627861023,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4850
+    },
+    {
+      "epoch": 5.320197044334975,
+      "grad_norm": 1.390471339225769,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 4860
+    },
+    {
+      "epoch": 5.331143951833607,
+      "grad_norm": 1.129770278930664,
+      "learning_rate": 0.0002,
+      "loss": 0.4098,
+      "step": 4870
+    },
+    {
+      "epoch": 5.342090859332239,
+      "grad_norm": 1.1461067199707031,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 4880
+    },
+    {
+      "epoch": 5.35303776683087,
+      "grad_norm": 1.3587424755096436,
+      "learning_rate": 0.0002,
+      "loss": 0.288,
+      "step": 4890
+    },
+    {
+      "epoch": 5.363984674329502,
+      "grad_norm": 1.6897879838943481,
+      "learning_rate": 0.0002,
+      "loss": 0.3604,
+      "step": 4900
+    },
+    {
+      "epoch": 5.374931581828133,
+      "grad_norm": 0.9298055768013,
+      "learning_rate": 0.0002,
+      "loss": 0.3887,
+      "step": 4910
+    },
+    {
+      "epoch": 5.385878489326765,
+      "grad_norm": 1.0006917715072632,
+      "learning_rate": 0.0002,
+      "loss": 0.3371,
+      "step": 4920
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 1.232581377029419,
+      "learning_rate": 0.0002,
+      "loss": 0.3992,
+      "step": 4930
+    },
+    {
+      "epoch": 5.407772304324029,
+      "grad_norm": 1.0822620391845703,
+      "learning_rate": 0.0002,
+      "loss": 0.3456,
+      "step": 4940
+    },
+    {
+      "epoch": 5.41871921182266,
+      "grad_norm": 1.3648720979690552,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4950
+    },
+    {
+      "epoch": 5.429666119321292,
+      "grad_norm": 1.3220354318618774,
+      "learning_rate": 0.0002,
+      "loss": 0.3959,
+      "step": 4960
+    },
+    {
+      "epoch": 5.440613026819923,
+      "grad_norm": 1.1106271743774414,
+      "learning_rate": 0.0002,
+      "loss": 0.3278,
+      "step": 4970
+    },
+    {
+      "epoch": 5.451559934318555,
+      "grad_norm": 1.6058908700942993,
+      "learning_rate": 0.0002,
+      "loss": 0.3812,
+      "step": 4980
+    },
+    {
+      "epoch": 5.462506841817187,
+      "grad_norm": 1.1065930128097534,
+      "learning_rate": 0.0002,
+      "loss": 0.3905,
+      "step": 4990
+    },
+    {
+      "epoch": 5.473453749315818,
+      "grad_norm": 1.3896466493606567,
+      "learning_rate": 0.0002,
+      "loss": 0.4058,
+      "step": 5000
+    },
+    {
+      "epoch": 5.48440065681445,
+      "grad_norm": 1.0437148809432983,
+      "learning_rate": 0.0002,
+      "loss": 0.4122,
+      "step": 5010
+    },
+    {
+      "epoch": 5.495347564313081,
+      "grad_norm": 1.2347718477249146,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 5020
+    },
+    {
+      "epoch": 5.506294471811713,
+      "grad_norm": 1.1174284219741821,
+      "learning_rate": 0.0002,
+      "loss": 0.3586,
+      "step": 5030
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2580941915512085,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 5040
+    },
+    {
+      "epoch": 5.528188286808977,
+      "grad_norm": 1.451090931892395,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 5050
+    },
+    {
+      "epoch": 5.539135194307608,
+      "grad_norm": 1.4688365459442139,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 5060
+    },
+    {
+      "epoch": 5.55008210180624,
+      "grad_norm": 1.1625734567642212,
+      "learning_rate": 0.0002,
+      "loss": 0.4431,
+      "step": 5070
+    },
+    {
+      "epoch": 5.561029009304871,
+      "grad_norm": 0.9332265257835388,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 5080
+    },
+    {
+      "epoch": 5.571975916803503,
+      "grad_norm": 1.5635273456573486,
+      "learning_rate": 0.0002,
+      "loss": 0.4,
+      "step": 5090
+    },
+    {
+      "epoch": 5.582922824302135,
+      "grad_norm": 1.3420509099960327,
+      "learning_rate": 0.0002,
+      "loss": 0.3651,
+      "step": 5100
+    },
+    {
+      "epoch": 5.593869731800766,
+      "grad_norm": 1.5826557874679565,
+      "learning_rate": 0.0002,
+      "loss": 0.3717,
+      "step": 5110
+    },
+    {
+      "epoch": 5.604816639299398,
+      "grad_norm": 1.5737065076828003,
+      "learning_rate": 0.0002,
+      "loss": 0.4256,
+      "step": 5120
+    },
+    {
+      "epoch": 5.615763546798029,
+      "grad_norm": 1.3812499046325684,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 5130
+    },
+    {
+      "epoch": 5.626710454296661,
+      "grad_norm": 1.362833023071289,
+      "learning_rate": 0.0002,
+      "loss": 0.3891,
+      "step": 5140
+    },
+    {
+      "epoch": 5.637657361795293,
+      "grad_norm": 1.7667874097824097,
+      "learning_rate": 0.0002,
+      "loss": 0.455,
+      "step": 5150
+    },
+    {
+      "epoch": 5.648604269293925,
+      "grad_norm": 1.2661789655685425,
+      "learning_rate": 0.0002,
+      "loss": 0.4264,
+      "step": 5160
+    },
+    {
+      "epoch": 5.659551176792556,
+      "grad_norm": 1.2076870203018188,
+      "learning_rate": 0.0002,
+      "loss": 0.3261,
+      "step": 5170
+    },
+    {
+      "epoch": 5.670498084291188,
+      "grad_norm": 1.2431524991989136,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 5180
+    },
+    {
+      "epoch": 5.681444991789819,
+      "grad_norm": 1.2216639518737793,
+      "learning_rate": 0.0002,
+      "loss": 0.4092,
+      "step": 5190
+    },
+    {
+      "epoch": 5.692391899288451,
+      "grad_norm": 0.9259352684020996,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5200
+    },
+    {
+      "epoch": 5.703338806787083,
+      "grad_norm": 1.7929338216781616,
+      "learning_rate": 0.0002,
+      "loss": 0.3875,
+      "step": 5210
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.4048460721969604,
+      "learning_rate": 0.0002,
+      "loss": 0.4424,
+      "step": 5220
+    },
+    {
+      "epoch": 5.725232621784346,
+      "grad_norm": 1.306874394416809,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 5230
+    },
+    {
+      "epoch": 5.736179529282977,
+      "grad_norm": 1.3137940168380737,
+      "learning_rate": 0.0002,
+      "loss": 0.3889,
+      "step": 5240
+    },
+    {
+      "epoch": 5.747126436781609,
+      "grad_norm": 1.1376476287841797,
+      "learning_rate": 0.0002,
+      "loss": 0.4804,
+      "step": 5250
+    },
+    {
+      "epoch": 5.758073344280241,
+      "grad_norm": 1.450939416885376,
+      "learning_rate": 0.0002,
+      "loss": 0.377,
+      "step": 5260
+    },
+    {
+      "epoch": 5.769020251778873,
+      "grad_norm": 0.983195960521698,
+      "learning_rate": 0.0002,
+      "loss": 0.4732,
+      "step": 5270
+    },
+    {
+      "epoch": 5.779967159277504,
+      "grad_norm": 1.66558837890625,
+      "learning_rate": 0.0002,
+      "loss": 0.4041,
+      "step": 5280
+    },
+    {
+      "epoch": 5.790914066776136,
+      "grad_norm": 0.9789204597473145,
+      "learning_rate": 0.0002,
+      "loss": 0.3643,
+      "step": 5290
+    },
+    {
+      "epoch": 5.801860974274767,
+      "grad_norm": 1.2110556364059448,
+      "learning_rate": 0.0002,
+      "loss": 0.3776,
+      "step": 5300
+    },
+    {
+      "epoch": 5.812807881773399,
+      "grad_norm": 1.3799304962158203,
+      "learning_rate": 0.0002,
+      "loss": 0.4049,
+      "step": 5310
+    },
+    {
+      "epoch": 5.823754789272031,
+      "grad_norm": 1.0570626258850098,
+      "learning_rate": 0.0002,
+      "loss": 0.4362,
+      "step": 5320
+    },
+    {
+      "epoch": 5.834701696770662,
+      "grad_norm": 1.4654436111450195,
+      "learning_rate": 0.0002,
+      "loss": 0.4716,
+      "step": 5330
+    },
+    {
+      "epoch": 5.845648604269294,
+      "grad_norm": 1.5216940641403198,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 5340
+    },
+    {
+      "epoch": 5.856595511767925,
+      "grad_norm": 1.018646001815796,
+      "learning_rate": 0.0002,
+      "loss": 0.3848,
+      "step": 5350
+    },
+    {
+      "epoch": 5.867542419266557,
+      "grad_norm": 1.028951644897461,
+      "learning_rate": 0.0002,
+      "loss": 0.3705,
+      "step": 5360
+    },
+    {
+      "epoch": 5.878489326765189,
+      "grad_norm": 2.571263313293457,
+      "learning_rate": 0.0002,
+      "loss": 0.4213,
+      "step": 5370
+    },
+    {
+      "epoch": 5.889436234263821,
+      "grad_norm": 1.3323984146118164,
+      "learning_rate": 0.0002,
+      "loss": 0.3647,
+      "step": 5380
+    },
+    {
+      "epoch": 5.900383141762452,
+      "grad_norm": 1.4317777156829834,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 5390
+    },
+    {
+      "epoch": 5.911330049261084,
+      "grad_norm": 1.4289140701293945,
+      "learning_rate": 0.0002,
+      "loss": 0.4254,
+      "step": 5400
+    },
+    {
+      "epoch": 5.922276956759715,
+      "grad_norm": 1.3130780458450317,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 5410
+    },
+    {
+      "epoch": 5.933223864258347,
+      "grad_norm": 1.3979902267456055,
+      "learning_rate": 0.0002,
+      "loss": 0.4025,
+      "step": 5420
+    },
+    {
+      "epoch": 5.944170771756979,
+      "grad_norm": 1.1827352046966553,
+      "learning_rate": 0.0002,
+      "loss": 0.3997,
+      "step": 5430
+    },
+    {
+      "epoch": 5.95511767925561,
+      "grad_norm": 1.1672080755233765,
+      "learning_rate": 0.0002,
+      "loss": 0.4163,
+      "step": 5440
+    },
+    {
+      "epoch": 5.966064586754242,
+      "grad_norm": 1.0949620008468628,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 5450
+    },
+    {
+      "epoch": 5.977011494252873,
+      "grad_norm": 1.3183925151824951,
+      "learning_rate": 0.0002,
+      "loss": 0.4219,
+      "step": 5460
+    },
+    {
+      "epoch": 5.987958401751505,
+      "grad_norm": 1.096198320388794,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5470
+    },
+    {
+      "epoch": 5.998905309250137,
+      "grad_norm": 1.2601423263549805,
+      "learning_rate": 0.0002,
+      "loss": 0.3886,
+      "step": 5480
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.611358880996704,
+      "eval_runtime": 46.0638,
+      "eval_samples_per_second": 9.465,
+      "eval_steps_per_second": 1.194,
+      "step": 5481
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 2.8152468978794496e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-5481/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..963c74622eea838273ac83d283aa57de83708ce3
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb59499751e431b269a158e9577079b7719b9d361e6a6d75b69339e4bd45608f
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6507afe6b978b0d95da2941b604aef718ef82e8a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe43d0b9cd92304f1825951897180e94734f0f9154fb25ea7086284cba22f982
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..257580bde1c6d381934a137c263206d24659d999
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:701709a539d7e9af65f1575bb9d1da237851fa2992973f27c5eb8eacb440fc72
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e23496075950221d743aae990e9f723697f33446
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1ec7cfbdc44640c90cecdfe9354667f6017ab26314148e5d93c1ae755b81ed
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..68803fbe4da3716ec3d1df6fae2887a2e349a160
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/trainer_state.json
@@ -0,0 +1,4562 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 6.999452654625069,
+  "eval_steps": 10,
+  "global_step": 6394,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    },
+    {
+      "epoch": 3.0103995621237,
+      "grad_norm": 0.6384502053260803,
+      "learning_rate": 0.0002,
+      "loss": 0.6443,
+      "step": 2750
+    },
+    {
+      "epoch": 3.0213464696223316,
+      "grad_norm": 0.9928722381591797,
+      "learning_rate": 0.0002,
+      "loss": 0.7123,
+      "step": 2760
+    },
+    {
+      "epoch": 3.0322933771209635,
+      "grad_norm": 0.7813051342964172,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 2770
+    },
+    {
+      "epoch": 3.043240284619595,
+      "grad_norm": 1.0202556848526,
+      "learning_rate": 0.0002,
+      "loss": 0.6042,
+      "step": 2780
+    },
+    {
+      "epoch": 3.0541871921182264,
+      "grad_norm": 0.7581062316894531,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 2790
+    },
+    {
+      "epoch": 3.0651340996168583,
+      "grad_norm": 0.6252710223197937,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 2800
+    },
+    {
+      "epoch": 3.07608100711549,
+      "grad_norm": 0.7738662958145142,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 2810
+    },
+    {
+      "epoch": 3.0870279146141213,
+      "grad_norm": 0.7381885051727295,
+      "learning_rate": 0.0002,
+      "loss": 0.627,
+      "step": 2820
+    },
+    {
+      "epoch": 3.097974822112753,
+      "grad_norm": 0.9197564721107483,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 2830
+    },
+    {
+      "epoch": 3.1089217296113847,
+      "grad_norm": 1.000976800918579,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 2840
+    },
+    {
+      "epoch": 3.1198686371100166,
+      "grad_norm": 0.7559131383895874,
+      "learning_rate": 0.0002,
+      "loss": 0.6631,
+      "step": 2850
+    },
+    {
+      "epoch": 3.130815544608648,
+      "grad_norm": 0.7213780879974365,
+      "learning_rate": 0.0002,
+      "loss": 0.6252,
+      "step": 2860
+    },
+    {
+      "epoch": 3.1417624521072796,
+      "grad_norm": 0.945939838886261,
+      "learning_rate": 0.0002,
+      "loss": 0.6501,
+      "step": 2870
+    },
+    {
+      "epoch": 3.1527093596059115,
+      "grad_norm": 0.7277454137802124,
+      "learning_rate": 0.0002,
+      "loss": 0.6129,
+      "step": 2880
+    },
+    {
+      "epoch": 3.163656267104543,
+      "grad_norm": 0.762026846408844,
+      "learning_rate": 0.0002,
+      "loss": 0.6423,
+      "step": 2890
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6471221446990967,
+      "learning_rate": 0.0002,
+      "loss": 0.5332,
+      "step": 2900
+    },
+    {
+      "epoch": 3.1855500821018063,
+      "grad_norm": 0.6018978357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 2910
+    },
+    {
+      "epoch": 3.196496989600438,
+      "grad_norm": 0.8607320785522461,
+      "learning_rate": 0.0002,
+      "loss": 0.7274,
+      "step": 2920
+    },
+    {
+      "epoch": 3.2074438970990693,
+      "grad_norm": 0.8854126334190369,
+      "learning_rate": 0.0002,
+      "loss": 0.6139,
+      "step": 2930
+    },
+    {
+      "epoch": 3.218390804597701,
+      "grad_norm": 0.6620870232582092,
+      "learning_rate": 0.0002,
+      "loss": 0.6485,
+      "step": 2940
+    },
+    {
+      "epoch": 3.2293377120963327,
+      "grad_norm": 0.7377511858940125,
+      "learning_rate": 0.0002,
+      "loss": 0.6969,
+      "step": 2950
+    },
+    {
+      "epoch": 3.2402846195949646,
+      "grad_norm": 0.7803301811218262,
+      "learning_rate": 0.0002,
+      "loss": 0.6798,
+      "step": 2960
+    },
+    {
+      "epoch": 3.251231527093596,
+      "grad_norm": 0.834061861038208,
+      "learning_rate": 0.0002,
+      "loss": 0.6697,
+      "step": 2970
+    },
+    {
+      "epoch": 3.2621784345922276,
+      "grad_norm": 0.8496041893959045,
+      "learning_rate": 0.0002,
+      "loss": 0.6894,
+      "step": 2980
+    },
+    {
+      "epoch": 3.2731253420908595,
+      "grad_norm": 0.7967984676361084,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 2990
+    },
+    {
+      "epoch": 3.284072249589491,
+      "grad_norm": 1.0207016468048096,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 3000
+    },
+    {
+      "epoch": 3.2950191570881224,
+      "grad_norm": 0.7855771780014038,
+      "learning_rate": 0.0002,
+      "loss": 0.6586,
+      "step": 3010
+    },
+    {
+      "epoch": 3.3059660645867543,
+      "grad_norm": 0.9427546858787537,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 3020
+    },
+    {
+      "epoch": 3.316912972085386,
+      "grad_norm": 0.823542594909668,
+      "learning_rate": 0.0002,
+      "loss": 0.6277,
+      "step": 3030
+    },
+    {
+      "epoch": 3.3278598795840173,
+      "grad_norm": 0.9826635122299194,
+      "learning_rate": 0.0002,
+      "loss": 0.7109,
+      "step": 3040
+    },
+    {
+      "epoch": 3.338806787082649,
+      "grad_norm": 0.7259827852249146,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 3050
+    },
+    {
+      "epoch": 3.3497536945812807,
+      "grad_norm": 0.7774739861488342,
+      "learning_rate": 0.0002,
+      "loss": 0.653,
+      "step": 3060
+    },
+    {
+      "epoch": 3.3607006020799126,
+      "grad_norm": 0.7394293546676636,
+      "learning_rate": 0.0002,
+      "loss": 0.7529,
+      "step": 3070
+    },
+    {
+      "epoch": 3.371647509578544,
+      "grad_norm": 0.9017578959465027,
+      "learning_rate": 0.0002,
+      "loss": 0.5987,
+      "step": 3080
+    },
+    {
+      "epoch": 3.3825944170771756,
+      "grad_norm": 0.7451054453849792,
+      "learning_rate": 0.0002,
+      "loss": 0.6953,
+      "step": 3090
+    },
+    {
+      "epoch": 3.3935413245758075,
+      "grad_norm": 0.7321506142616272,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 3100
+    },
+    {
+      "epoch": 3.404488232074439,
+      "grad_norm": 0.6721828579902649,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3110
+    },
+    {
+      "epoch": 3.4154351395730704,
+      "grad_norm": 0.774022102355957,
+      "learning_rate": 0.0002,
+      "loss": 0.6559,
+      "step": 3120
+    },
+    {
+      "epoch": 3.4263820470717024,
+      "grad_norm": 0.9143537282943726,
+      "learning_rate": 0.0002,
+      "loss": 0.7449,
+      "step": 3130
+    },
+    {
+      "epoch": 3.437328954570334,
+      "grad_norm": 1.226087212562561,
+      "learning_rate": 0.0002,
+      "loss": 0.6899,
+      "step": 3140
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7545496225357056,
+      "learning_rate": 0.0002,
+      "loss": 0.6719,
+      "step": 3150
+    },
+    {
+      "epoch": 3.4592227695675972,
+      "grad_norm": 0.6515635848045349,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 3160
+    },
+    {
+      "epoch": 3.4701696770662287,
+      "grad_norm": 0.9297090172767639,
+      "learning_rate": 0.0002,
+      "loss": 0.6926,
+      "step": 3170
+    },
+    {
+      "epoch": 3.4811165845648606,
+      "grad_norm": 1.0130730867385864,
+      "learning_rate": 0.0002,
+      "loss": 0.6071,
+      "step": 3180
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.7654589414596558,
+      "learning_rate": 0.0002,
+      "loss": 0.5959,
+      "step": 3190
+    },
+    {
+      "epoch": 3.5030103995621236,
+      "grad_norm": 0.9954977631568909,
+      "learning_rate": 0.0002,
+      "loss": 0.7401,
+      "step": 3200
+    },
+    {
+      "epoch": 3.5139573070607555,
+      "grad_norm": 0.6027487516403198,
+      "learning_rate": 0.0002,
+      "loss": 0.6661,
+      "step": 3210
+    },
+    {
+      "epoch": 3.524904214559387,
+      "grad_norm": 0.741770327091217,
+      "learning_rate": 0.0002,
+      "loss": 0.6963,
+      "step": 3220
+    },
+    {
+      "epoch": 3.535851122058019,
+      "grad_norm": 1.0534909963607788,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 3230
+    },
+    {
+      "epoch": 3.5467980295566504,
+      "grad_norm": 0.937772274017334,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 3240
+    },
+    {
+      "epoch": 3.557744937055282,
+      "grad_norm": 0.8504213690757751,
+      "learning_rate": 0.0002,
+      "loss": 0.6681,
+      "step": 3250
+    },
+    {
+      "epoch": 3.5686918445539133,
+      "grad_norm": 0.7755007147789001,
+      "learning_rate": 0.0002,
+      "loss": 0.6436,
+      "step": 3260
+    },
+    {
+      "epoch": 3.5796387520525452,
+      "grad_norm": 1.0193358659744263,
+      "learning_rate": 0.0002,
+      "loss": 0.6213,
+      "step": 3270
+    },
+    {
+      "epoch": 3.5905856595511767,
+      "grad_norm": 0.8440536856651306,
+      "learning_rate": 0.0002,
+      "loss": 0.671,
+      "step": 3280
+    },
+    {
+      "epoch": 3.6015325670498086,
+      "grad_norm": 0.6195939183235168,
+      "learning_rate": 0.0002,
+      "loss": 0.6859,
+      "step": 3290
+    },
+    {
+      "epoch": 3.61247947454844,
+      "grad_norm": 0.8608590960502625,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3300
+    },
+    {
+      "epoch": 3.6234263820470716,
+      "grad_norm": 0.6772327423095703,
+      "learning_rate": 0.0002,
+      "loss": 0.7301,
+      "step": 3310
+    },
+    {
+      "epoch": 3.6343732895457035,
+      "grad_norm": 0.8031839728355408,
+      "learning_rate": 0.0002,
+      "loss": 0.6298,
+      "step": 3320
+    },
+    {
+      "epoch": 3.645320197044335,
+      "grad_norm": 0.6080502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.7041,
+      "step": 3330
+    },
+    {
+      "epoch": 3.656267104542967,
+      "grad_norm": 0.8007240891456604,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 3340
+    },
+    {
+      "epoch": 3.6672140120415984,
+      "grad_norm": 0.8060704469680786,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3350
+    },
+    {
+      "epoch": 3.67816091954023,
+      "grad_norm": 0.7547586560249329,
+      "learning_rate": 0.0002,
+      "loss": 0.6304,
+      "step": 3360
+    },
+    {
+      "epoch": 3.6891078270388613,
+      "grad_norm": 0.686851978302002,
+      "learning_rate": 0.0002,
+      "loss": 0.7066,
+      "step": 3370
+    },
+    {
+      "epoch": 3.7000547345374932,
+      "grad_norm": 0.9429075717926025,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 3380
+    },
+    {
+      "epoch": 3.7110016420361247,
+      "grad_norm": 0.7283591032028198,
+      "learning_rate": 0.0002,
+      "loss": 0.6673,
+      "step": 3390
+    },
+    {
+      "epoch": 3.7219485495347566,
+      "grad_norm": 0.8323085904121399,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 3400
+    },
+    {
+      "epoch": 3.732895457033388,
+      "grad_norm": 0.8529590964317322,
+      "learning_rate": 0.0002,
+      "loss": 0.7779,
+      "step": 3410
+    },
+    {
+      "epoch": 3.7438423645320196,
+      "grad_norm": 0.731752872467041,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3420
+    },
+    {
+      "epoch": 3.7547892720306515,
+      "grad_norm": 0.8572278618812561,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3430
+    },
+    {
+      "epoch": 3.765736179529283,
+      "grad_norm": 0.7408691048622131,
+      "learning_rate": 0.0002,
+      "loss": 0.6215,
+      "step": 3440
+    },
+    {
+      "epoch": 3.776683087027915,
+      "grad_norm": 0.7470445036888123,
+      "learning_rate": 0.0002,
+      "loss": 0.622,
+      "step": 3450
+    },
+    {
+      "epoch": 3.7876299945265464,
+      "grad_norm": 0.6806244254112244,
+      "learning_rate": 0.0002,
+      "loss": 0.7241,
+      "step": 3460
+    },
+    {
+      "epoch": 3.798576902025178,
+      "grad_norm": 0.9129069447517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7739,
+      "step": 3470
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.8717501759529114,
+      "learning_rate": 0.0002,
+      "loss": 0.6826,
+      "step": 3480
+    },
+    {
+      "epoch": 3.8204707170224412,
+      "grad_norm": 0.6761979460716248,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 3490
+    },
+    {
+      "epoch": 3.8314176245210727,
+      "grad_norm": 1.0054380893707275,
+      "learning_rate": 0.0002,
+      "loss": 0.7601,
+      "step": 3500
+    },
+    {
+      "epoch": 3.8423645320197046,
+      "grad_norm": 1.1224009990692139,
+      "learning_rate": 0.0002,
+      "loss": 0.623,
+      "step": 3510
+    },
+    {
+      "epoch": 3.853311439518336,
+      "grad_norm": 0.8997692465782166,
+      "learning_rate": 0.0002,
+      "loss": 0.6918,
+      "step": 3520
+    },
+    {
+      "epoch": 3.8642583470169676,
+      "grad_norm": 1.0086902379989624,
+      "learning_rate": 0.0002,
+      "loss": 0.6357,
+      "step": 3530
+    },
+    {
+      "epoch": 3.8752052545155995,
+      "grad_norm": 0.772739589214325,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 3540
+    },
+    {
+      "epoch": 3.886152162014231,
+      "grad_norm": 1.211774230003357,
+      "learning_rate": 0.0002,
+      "loss": 0.7423,
+      "step": 3550
+    },
+    {
+      "epoch": 3.897099069512863,
+      "grad_norm": 0.9572356939315796,
+      "learning_rate": 0.0002,
+      "loss": 0.7321,
+      "step": 3560
+    },
+    {
+      "epoch": 3.9080459770114944,
+      "grad_norm": 0.7887842655181885,
+      "learning_rate": 0.0002,
+      "loss": 0.6836,
+      "step": 3570
+    },
+    {
+      "epoch": 3.918992884510126,
+      "grad_norm": 0.7308389544487,
+      "learning_rate": 0.0002,
+      "loss": 0.7576,
+      "step": 3580
+    },
+    {
+      "epoch": 3.9299397920087573,
+      "grad_norm": 1.0182650089263916,
+      "learning_rate": 0.0002,
+      "loss": 0.6001,
+      "step": 3590
+    },
+    {
+      "epoch": 3.9408866995073892,
+      "grad_norm": 0.8000147342681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 3600
+    },
+    {
+      "epoch": 3.9518336070060207,
+      "grad_norm": 0.7385728359222412,
+      "learning_rate": 0.0002,
+      "loss": 0.6244,
+      "step": 3610
+    },
+    {
+      "epoch": 3.9627805145046526,
+      "grad_norm": 0.9233261942863464,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 3620
+    },
+    {
+      "epoch": 3.973727422003284,
+      "grad_norm": 0.8486751914024353,
+      "learning_rate": 0.0002,
+      "loss": 0.6508,
+      "step": 3630
+    },
+    {
+      "epoch": 3.9846743295019156,
+      "grad_norm": 0.7593663334846497,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3640
+    },
+    {
+      "epoch": 3.9956212370005475,
+      "grad_norm": 0.7885415554046631,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 3650
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.250312328338623,
+      "eval_runtime": 46.0842,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.193,
+      "step": 3654
+    },
+    {
+      "epoch": 4.006568144499179,
+      "grad_norm": 0.6591703295707703,
+      "learning_rate": 0.0002,
+      "loss": 0.5547,
+      "step": 3660
+    },
+    {
+      "epoch": 4.017515051997811,
+      "grad_norm": 1.36927330493927,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 3670
+    },
+    {
+      "epoch": 4.028461959496442,
+      "grad_norm": 0.8106328845024109,
+      "learning_rate": 0.0002,
+      "loss": 0.4466,
+      "step": 3680
+    },
+    {
+      "epoch": 4.039408866995074,
+      "grad_norm": 0.7592712044715881,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 3690
+    },
+    {
+      "epoch": 4.050355774493705,
+      "grad_norm": 0.9518909454345703,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3700
+    },
+    {
+      "epoch": 4.061302681992337,
+      "grad_norm": 0.7805967330932617,
+      "learning_rate": 0.0002,
+      "loss": 0.4638,
+      "step": 3710
+    },
+    {
+      "epoch": 4.072249589490969,
+      "grad_norm": 1.3146334886550903,
+      "learning_rate": 0.0002,
+      "loss": 0.4556,
+      "step": 3720
+    },
+    {
+      "epoch": 4.083196496989601,
+      "grad_norm": 1.1611138582229614,
+      "learning_rate": 0.0002,
+      "loss": 0.5635,
+      "step": 3730
+    },
+    {
+      "epoch": 4.094143404488232,
+      "grad_norm": 0.8173232078552246,
+      "learning_rate": 0.0002,
+      "loss": 0.3845,
+      "step": 3740
+    },
+    {
+      "epoch": 4.105090311986864,
+      "grad_norm": 0.7848323583602905,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 3750
+    },
+    {
+      "epoch": 4.116037219485495,
+      "grad_norm": 1.3183201551437378,
+      "learning_rate": 0.0002,
+      "loss": 0.4519,
+      "step": 3760
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 1.1936529874801636,
+      "learning_rate": 0.0002,
+      "loss": 0.5083,
+      "step": 3770
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1078993082046509,
+      "learning_rate": 0.0002,
+      "loss": 0.5208,
+      "step": 3780
+    },
+    {
+      "epoch": 4.14887794198139,
+      "grad_norm": 1.107743263244629,
+      "learning_rate": 0.0002,
+      "loss": 0.5928,
+      "step": 3790
+    },
+    {
+      "epoch": 4.159824849480022,
+      "grad_norm": 0.7801875472068787,
+      "learning_rate": 0.0002,
+      "loss": 0.5112,
+      "step": 3800
+    },
+    {
+      "epoch": 4.170771756978653,
+      "grad_norm": 1.1328117847442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4896,
+      "step": 3810
+    },
+    {
+      "epoch": 4.181718664477285,
+      "grad_norm": 1.4232193231582642,
+      "learning_rate": 0.0002,
+      "loss": 0.5645,
+      "step": 3820
+    },
+    {
+      "epoch": 4.192665571975917,
+      "grad_norm": 1.557416558265686,
+      "learning_rate": 0.0002,
+      "loss": 0.5049,
+      "step": 3830
+    },
+    {
+      "epoch": 4.203612479474549,
+      "grad_norm": 1.042923092842102,
+      "learning_rate": 0.0002,
+      "loss": 0.4863,
+      "step": 3840
+    },
+    {
+      "epoch": 4.21455938697318,
+      "grad_norm": 1.1801949739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 3850
+    },
+    {
+      "epoch": 4.225506294471812,
+      "grad_norm": 0.9273753762245178,
+      "learning_rate": 0.0002,
+      "loss": 0.5063,
+      "step": 3860
+    },
+    {
+      "epoch": 4.236453201970443,
+      "grad_norm": 0.7681763768196106,
+      "learning_rate": 0.0002,
+      "loss": 0.5542,
+      "step": 3870
+    },
+    {
+      "epoch": 4.2474001094690745,
+      "grad_norm": 0.9840841293334961,
+      "learning_rate": 0.0002,
+      "loss": 0.5971,
+      "step": 3880
+    },
+    {
+      "epoch": 4.258347016967707,
+      "grad_norm": 1.0290725231170654,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 3890
+    },
+    {
+      "epoch": 4.269293924466338,
+      "grad_norm": 0.8059597611427307,
+      "learning_rate": 0.0002,
+      "loss": 0.4288,
+      "step": 3900
+    },
+    {
+      "epoch": 4.28024083196497,
+      "grad_norm": 0.9847467541694641,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3910
+    },
+    {
+      "epoch": 4.291187739463601,
+      "grad_norm": 1.344044804573059,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 3920
+    },
+    {
+      "epoch": 4.302134646962233,
+      "grad_norm": 0.9174224138259888,
+      "learning_rate": 0.0002,
+      "loss": 0.4966,
+      "step": 3930
+    },
+    {
+      "epoch": 4.313081554460865,
+      "grad_norm": 1.1199711561203003,
+      "learning_rate": 0.0002,
+      "loss": 0.4944,
+      "step": 3940
+    },
+    {
+      "epoch": 4.324028461959497,
+      "grad_norm": 1.0120296478271484,
+      "learning_rate": 0.0002,
+      "loss": 0.4641,
+      "step": 3950
+    },
+    {
+      "epoch": 4.334975369458128,
+      "grad_norm": 1.091811180114746,
+      "learning_rate": 0.0002,
+      "loss": 0.4723,
+      "step": 3960
+    },
+    {
+      "epoch": 4.34592227695676,
+      "grad_norm": 1.0332133769989014,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 3970
+    },
+    {
+      "epoch": 4.356869184455391,
+      "grad_norm": 1.0785295963287354,
+      "learning_rate": 0.0002,
+      "loss": 0.4646,
+      "step": 3980
+    },
+    {
+      "epoch": 4.3678160919540225,
+      "grad_norm": 1.0506969690322876,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 3990
+    },
+    {
+      "epoch": 4.378762999452655,
+      "grad_norm": 1.047560691833496,
+      "learning_rate": 0.0002,
+      "loss": 0.4776,
+      "step": 4000
+    },
+    {
+      "epoch": 4.389709906951286,
+      "grad_norm": 0.9348800778388977,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 4010
+    },
+    {
+      "epoch": 4.400656814449918,
+      "grad_norm": 1.1563059091567993,
+      "learning_rate": 0.0002,
+      "loss": 0.5333,
+      "step": 4020
+    },
+    {
+      "epoch": 4.411603721948549,
+      "grad_norm": 1.001470923423767,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 4030
+    },
+    {
+      "epoch": 4.422550629447181,
+      "grad_norm": 1.309012532234192,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 4040
+    },
+    {
+      "epoch": 4.433497536945813,
+      "grad_norm": 0.7338925004005432,
+      "learning_rate": 0.0002,
+      "loss": 0.5078,
+      "step": 4050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 1.0398834943771362,
+      "learning_rate": 0.0002,
+      "loss": 0.4632,
+      "step": 4060
+    },
+    {
+      "epoch": 4.455391351943076,
+      "grad_norm": 0.9728689193725586,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4070
+    },
+    {
+      "epoch": 4.466338259441708,
+      "grad_norm": 1.247475028038025,
+      "learning_rate": 0.0002,
+      "loss": 0.4778,
+      "step": 4080
+    },
+    {
+      "epoch": 4.477285166940339,
+      "grad_norm": 1.1084578037261963,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 4090
+    },
+    {
+      "epoch": 4.4882320744389705,
+      "grad_norm": 1.1619318723678589,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 4100
+    },
+    {
+      "epoch": 4.499178981937603,
+      "grad_norm": 1.3456498384475708,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4110
+    },
+    {
+      "epoch": 4.510125889436234,
+      "grad_norm": 0.9372991323471069,
+      "learning_rate": 0.0002,
+      "loss": 0.4876,
+      "step": 4120
+    },
+    {
+      "epoch": 4.521072796934866,
+      "grad_norm": 1.0071815252304077,
+      "learning_rate": 0.0002,
+      "loss": 0.5456,
+      "step": 4130
+    },
+    {
+      "epoch": 4.532019704433497,
+      "grad_norm": 1.190344214439392,
+      "learning_rate": 0.0002,
+      "loss": 0.5589,
+      "step": 4140
+    },
+    {
+      "epoch": 4.542966611932129,
+      "grad_norm": 0.9480887055397034,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 4150
+    },
+    {
+      "epoch": 4.553913519430761,
+      "grad_norm": 1.0252189636230469,
+      "learning_rate": 0.0002,
+      "loss": 0.5229,
+      "step": 4160
+    },
+    {
+      "epoch": 4.564860426929393,
+      "grad_norm": 0.7142013311386108,
+      "learning_rate": 0.0002,
+      "loss": 0.5253,
+      "step": 4170
+    },
+    {
+      "epoch": 4.575807334428024,
+      "grad_norm": 0.8937426805496216,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 4180
+    },
+    {
+      "epoch": 4.586754241926656,
+      "grad_norm": 0.8885005116462708,
+      "learning_rate": 0.0002,
+      "loss": 0.4773,
+      "step": 4190
+    },
+    {
+      "epoch": 4.597701149425287,
+      "grad_norm": 1.337663173675537,
+      "learning_rate": 0.0002,
+      "loss": 0.4858,
+      "step": 4200
+    },
+    {
+      "epoch": 4.6086480569239185,
+      "grad_norm": 1.0475375652313232,
+      "learning_rate": 0.0002,
+      "loss": 0.5247,
+      "step": 4210
+    },
+    {
+      "epoch": 4.619594964422551,
+      "grad_norm": 1.0081088542938232,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 4220
+    },
+    {
+      "epoch": 4.630541871921182,
+      "grad_norm": 0.7527595162391663,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 4230
+    },
+    {
+      "epoch": 4.641488779419814,
+      "grad_norm": 1.55559241771698,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4240
+    },
+    {
+      "epoch": 4.652435686918445,
+      "grad_norm": 0.7967379689216614,
+      "learning_rate": 0.0002,
+      "loss": 0.5468,
+      "step": 4250
+    },
+    {
+      "epoch": 4.663382594417077,
+      "grad_norm": 0.898368775844574,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 4260
+    },
+    {
+      "epoch": 4.674329501915709,
+      "grad_norm": 1.1940776109695435,
+      "learning_rate": 0.0002,
+      "loss": 0.4706,
+      "step": 4270
+    },
+    {
+      "epoch": 4.685276409414341,
+      "grad_norm": 1.1817092895507812,
+      "learning_rate": 0.0002,
+      "loss": 0.5121,
+      "step": 4280
+    },
+    {
+      "epoch": 4.696223316912972,
+      "grad_norm": 0.9041520357131958,
+      "learning_rate": 0.0002,
+      "loss": 0.5758,
+      "step": 4290
+    },
+    {
+      "epoch": 4.707170224411604,
+      "grad_norm": 1.1280102729797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5851,
+      "step": 4300
+    },
+    {
+      "epoch": 4.718117131910235,
+      "grad_norm": 1.357689619064331,
+      "learning_rate": 0.0002,
+      "loss": 0.4891,
+      "step": 4310
+    },
+    {
+      "epoch": 4.7290640394088665,
+      "grad_norm": 1.056633472442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4704,
+      "step": 4320
+    },
+    {
+      "epoch": 4.740010946907499,
+      "grad_norm": 1.6520427465438843,
+      "learning_rate": 0.0002,
+      "loss": 0.5488,
+      "step": 4330
+    },
+    {
+      "epoch": 4.75095785440613,
+      "grad_norm": 1.153200626373291,
+      "learning_rate": 0.0002,
+      "loss": 0.5131,
+      "step": 4340
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.9346241354942322,
+      "learning_rate": 0.0002,
+      "loss": 0.539,
+      "step": 4350
+    },
+    {
+      "epoch": 4.772851669403393,
+      "grad_norm": 0.8628455996513367,
+      "learning_rate": 0.0002,
+      "loss": 0.4941,
+      "step": 4360
+    },
+    {
+      "epoch": 4.783798576902025,
+      "grad_norm": 1.3843916654586792,
+      "learning_rate": 0.0002,
+      "loss": 0.5167,
+      "step": 4370
+    },
+    {
+      "epoch": 4.794745484400657,
+      "grad_norm": 1.035574197769165,
+      "learning_rate": 0.0002,
+      "loss": 0.4683,
+      "step": 4380
+    },
+    {
+      "epoch": 4.805692391899289,
+      "grad_norm": 1.1868361234664917,
+      "learning_rate": 0.0002,
+      "loss": 0.5162,
+      "step": 4390
+    },
+    {
+      "epoch": 4.81663929939792,
+      "grad_norm": 1.1307647228240967,
+      "learning_rate": 0.0002,
+      "loss": 0.534,
+      "step": 4400
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 0.9787724614143372,
+      "learning_rate": 0.0002,
+      "loss": 0.5567,
+      "step": 4410
+    },
+    {
+      "epoch": 4.838533114395183,
+      "grad_norm": 1.0473824739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.5185,
+      "step": 4420
+    },
+    {
+      "epoch": 4.8494800218938146,
+      "grad_norm": 1.069069504737854,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4430
+    },
+    {
+      "epoch": 4.860426929392447,
+      "grad_norm": 1.4305680990219116,
+      "learning_rate": 0.0002,
+      "loss": 0.5267,
+      "step": 4440
+    },
+    {
+      "epoch": 4.871373836891078,
+      "grad_norm": 1.3679203987121582,
+      "learning_rate": 0.0002,
+      "loss": 0.5947,
+      "step": 4450
+    },
+    {
+      "epoch": 4.88232074438971,
+      "grad_norm": 0.8997844457626343,
+      "learning_rate": 0.0002,
+      "loss": 0.5135,
+      "step": 4460
+    },
+    {
+      "epoch": 4.893267651888341,
+      "grad_norm": 1.2758110761642456,
+      "learning_rate": 0.0002,
+      "loss": 0.5312,
+      "step": 4470
+    },
+    {
+      "epoch": 4.904214559386973,
+      "grad_norm": 0.8819465637207031,
+      "learning_rate": 0.0002,
+      "loss": 0.4914,
+      "step": 4480
+    },
+    {
+      "epoch": 4.915161466885605,
+      "grad_norm": 1.08329439163208,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 4490
+    },
+    {
+      "epoch": 4.926108374384237,
+      "grad_norm": 1.083461046218872,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 4500
+    },
+    {
+      "epoch": 4.937055281882868,
+      "grad_norm": 1.2387723922729492,
+      "learning_rate": 0.0002,
+      "loss": 0.5433,
+      "step": 4510
+    },
+    {
+      "epoch": 4.9480021893815,
+      "grad_norm": 0.8262293934822083,
+      "learning_rate": 0.0002,
+      "loss": 0.5624,
+      "step": 4520
+    },
+    {
+      "epoch": 4.958949096880131,
+      "grad_norm": 1.2325191497802734,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 4530
+    },
+    {
+      "epoch": 4.9698960043787626,
+      "grad_norm": 1.024614930152893,
+      "learning_rate": 0.0002,
+      "loss": 0.5452,
+      "step": 4540
+    },
+    {
+      "epoch": 4.980842911877395,
+      "grad_norm": 1.3007521629333496,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 4550
+    },
+    {
+      "epoch": 4.991789819376026,
+      "grad_norm": 0.9823828339576721,
+      "learning_rate": 0.0002,
+      "loss": 0.4943,
+      "step": 4560
+    },
+    {
+      "epoch": 4.999452654625069,
+      "eval_loss": 1.3920727968215942,
+      "eval_runtime": 46.0764,
+      "eval_samples_per_second": 9.463,
+      "eval_steps_per_second": 1.194,
+      "step": 4567
+    },
+    {
+      "epoch": 5.002736726874658,
+      "grad_norm": 1.1478906869888306,
+      "learning_rate": 0.0002,
+      "loss": 0.545,
+      "step": 4570
+    },
+    {
+      "epoch": 5.013683634373289,
+      "grad_norm": 1.0533705949783325,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 4580
+    },
+    {
+      "epoch": 5.024630541871921,
+      "grad_norm": 1.268900752067566,
+      "learning_rate": 0.0002,
+      "loss": 0.3313,
+      "step": 4590
+    },
+    {
+      "epoch": 5.035577449370553,
+      "grad_norm": 1.222652554512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3482,
+      "step": 4600
+    },
+    {
+      "epoch": 5.046524356869185,
+      "grad_norm": 1.5093127489089966,
+      "learning_rate": 0.0002,
+      "loss": 0.3195,
+      "step": 4610
+    },
+    {
+      "epoch": 5.057471264367816,
+      "grad_norm": 1.2372499704360962,
+      "learning_rate": 0.0002,
+      "loss": 0.3569,
+      "step": 4620
+    },
+    {
+      "epoch": 5.068418171866448,
+      "grad_norm": 0.8422666192054749,
+      "learning_rate": 0.0002,
+      "loss": 0.3206,
+      "step": 4630
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 1.1451770067214966,
+      "learning_rate": 0.0002,
+      "loss": 0.3115,
+      "step": 4640
+    },
+    {
+      "epoch": 5.090311986863711,
+      "grad_norm": 1.2074557542800903,
+      "learning_rate": 0.0002,
+      "loss": 0.3305,
+      "step": 4650
+    },
+    {
+      "epoch": 5.101258894362343,
+      "grad_norm": 1.429150104522705,
+      "learning_rate": 0.0002,
+      "loss": 0.3012,
+      "step": 4660
+    },
+    {
+      "epoch": 5.112205801860974,
+      "grad_norm": 1.0353610515594482,
+      "learning_rate": 0.0002,
+      "loss": 0.3229,
+      "step": 4670
+    },
+    {
+      "epoch": 5.123152709359606,
+      "grad_norm": 1.2845979928970337,
+      "learning_rate": 0.0002,
+      "loss": 0.402,
+      "step": 4680
+    },
+    {
+      "epoch": 5.134099616858237,
+      "grad_norm": 1.3790186643600464,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 4690
+    },
+    {
+      "epoch": 5.145046524356869,
+      "grad_norm": 1.3182239532470703,
+      "learning_rate": 0.0002,
+      "loss": 0.2951,
+      "step": 4700
+    },
+    {
+      "epoch": 5.155993431855501,
+      "grad_norm": 1.5249626636505127,
+      "learning_rate": 0.0002,
+      "loss": 0.4074,
+      "step": 4710
+    },
+    {
+      "epoch": 5.166940339354133,
+      "grad_norm": 1.2492733001708984,
+      "learning_rate": 0.0002,
+      "loss": 0.3703,
+      "step": 4720
+    },
+    {
+      "epoch": 5.177887246852764,
+      "grad_norm": 1.4455480575561523,
+      "learning_rate": 0.0002,
+      "loss": 0.3411,
+      "step": 4730
+    },
+    {
+      "epoch": 5.188834154351396,
+      "grad_norm": 1.2191482782363892,
+      "learning_rate": 0.0002,
+      "loss": 0.3996,
+      "step": 4740
+    },
+    {
+      "epoch": 5.199781061850027,
+      "grad_norm": 1.4707951545715332,
+      "learning_rate": 0.0002,
+      "loss": 0.3785,
+      "step": 4750
+    },
+    {
+      "epoch": 5.210727969348659,
+      "grad_norm": 1.3473678827285767,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4760
+    },
+    {
+      "epoch": 5.221674876847291,
+      "grad_norm": 1.0479670763015747,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4770
+    },
+    {
+      "epoch": 5.232621784345922,
+      "grad_norm": 1.299096703529358,
+      "learning_rate": 0.0002,
+      "loss": 0.3976,
+      "step": 4780
+    },
+    {
+      "epoch": 5.243568691844554,
+      "grad_norm": 1.2820168733596802,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4790
+    },
+    {
+      "epoch": 5.254515599343185,
+      "grad_norm": 1.3818004131317139,
+      "learning_rate": 0.0002,
+      "loss": 0.3347,
+      "step": 4800
+    },
+    {
+      "epoch": 5.265462506841817,
+      "grad_norm": 1.2898736000061035,
+      "learning_rate": 0.0002,
+      "loss": 0.3761,
+      "step": 4810
+    },
+    {
+      "epoch": 5.276409414340449,
+      "grad_norm": 1.1761468648910522,
+      "learning_rate": 0.0002,
+      "loss": 0.3694,
+      "step": 4820
+    },
+    {
+      "epoch": 5.287356321839081,
+      "grad_norm": 1.7155952453613281,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4830
+    },
+    {
+      "epoch": 5.298303229337712,
+      "grad_norm": 0.9103642106056213,
+      "learning_rate": 0.0002,
+      "loss": 0.322,
+      "step": 4840
+    },
+    {
+      "epoch": 5.309250136836344,
+      "grad_norm": 1.013015627861023,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4850
+    },
+    {
+      "epoch": 5.320197044334975,
+      "grad_norm": 1.390471339225769,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 4860
+    },
+    {
+      "epoch": 5.331143951833607,
+      "grad_norm": 1.129770278930664,
+      "learning_rate": 0.0002,
+      "loss": 0.4098,
+      "step": 4870
+    },
+    {
+      "epoch": 5.342090859332239,
+      "grad_norm": 1.1461067199707031,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 4880
+    },
+    {
+      "epoch": 5.35303776683087,
+      "grad_norm": 1.3587424755096436,
+      "learning_rate": 0.0002,
+      "loss": 0.288,
+      "step": 4890
+    },
+    {
+      "epoch": 5.363984674329502,
+      "grad_norm": 1.6897879838943481,
+      "learning_rate": 0.0002,
+      "loss": 0.3604,
+      "step": 4900
+    },
+    {
+      "epoch": 5.374931581828133,
+      "grad_norm": 0.9298055768013,
+      "learning_rate": 0.0002,
+      "loss": 0.3887,
+      "step": 4910
+    },
+    {
+      "epoch": 5.385878489326765,
+      "grad_norm": 1.0006917715072632,
+      "learning_rate": 0.0002,
+      "loss": 0.3371,
+      "step": 4920
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 1.232581377029419,
+      "learning_rate": 0.0002,
+      "loss": 0.3992,
+      "step": 4930
+    },
+    {
+      "epoch": 5.407772304324029,
+      "grad_norm": 1.0822620391845703,
+      "learning_rate": 0.0002,
+      "loss": 0.3456,
+      "step": 4940
+    },
+    {
+      "epoch": 5.41871921182266,
+      "grad_norm": 1.3648720979690552,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4950
+    },
+    {
+      "epoch": 5.429666119321292,
+      "grad_norm": 1.3220354318618774,
+      "learning_rate": 0.0002,
+      "loss": 0.3959,
+      "step": 4960
+    },
+    {
+      "epoch": 5.440613026819923,
+      "grad_norm": 1.1106271743774414,
+      "learning_rate": 0.0002,
+      "loss": 0.3278,
+      "step": 4970
+    },
+    {
+      "epoch": 5.451559934318555,
+      "grad_norm": 1.6058908700942993,
+      "learning_rate": 0.0002,
+      "loss": 0.3812,
+      "step": 4980
+    },
+    {
+      "epoch": 5.462506841817187,
+      "grad_norm": 1.1065930128097534,
+      "learning_rate": 0.0002,
+      "loss": 0.3905,
+      "step": 4990
+    },
+    {
+      "epoch": 5.473453749315818,
+      "grad_norm": 1.3896466493606567,
+      "learning_rate": 0.0002,
+      "loss": 0.4058,
+      "step": 5000
+    },
+    {
+      "epoch": 5.48440065681445,
+      "grad_norm": 1.0437148809432983,
+      "learning_rate": 0.0002,
+      "loss": 0.4122,
+      "step": 5010
+    },
+    {
+      "epoch": 5.495347564313081,
+      "grad_norm": 1.2347718477249146,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 5020
+    },
+    {
+      "epoch": 5.506294471811713,
+      "grad_norm": 1.1174284219741821,
+      "learning_rate": 0.0002,
+      "loss": 0.3586,
+      "step": 5030
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2580941915512085,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 5040
+    },
+    {
+      "epoch": 5.528188286808977,
+      "grad_norm": 1.451090931892395,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 5050
+    },
+    {
+      "epoch": 5.539135194307608,
+      "grad_norm": 1.4688365459442139,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 5060
+    },
+    {
+      "epoch": 5.55008210180624,
+      "grad_norm": 1.1625734567642212,
+      "learning_rate": 0.0002,
+      "loss": 0.4431,
+      "step": 5070
+    },
+    {
+      "epoch": 5.561029009304871,
+      "grad_norm": 0.9332265257835388,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 5080
+    },
+    {
+      "epoch": 5.571975916803503,
+      "grad_norm": 1.5635273456573486,
+      "learning_rate": 0.0002,
+      "loss": 0.4,
+      "step": 5090
+    },
+    {
+      "epoch": 5.582922824302135,
+      "grad_norm": 1.3420509099960327,
+      "learning_rate": 0.0002,
+      "loss": 0.3651,
+      "step": 5100
+    },
+    {
+      "epoch": 5.593869731800766,
+      "grad_norm": 1.5826557874679565,
+      "learning_rate": 0.0002,
+      "loss": 0.3717,
+      "step": 5110
+    },
+    {
+      "epoch": 5.604816639299398,
+      "grad_norm": 1.5737065076828003,
+      "learning_rate": 0.0002,
+      "loss": 0.4256,
+      "step": 5120
+    },
+    {
+      "epoch": 5.615763546798029,
+      "grad_norm": 1.3812499046325684,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 5130
+    },
+    {
+      "epoch": 5.626710454296661,
+      "grad_norm": 1.362833023071289,
+      "learning_rate": 0.0002,
+      "loss": 0.3891,
+      "step": 5140
+    },
+    {
+      "epoch": 5.637657361795293,
+      "grad_norm": 1.7667874097824097,
+      "learning_rate": 0.0002,
+      "loss": 0.455,
+      "step": 5150
+    },
+    {
+      "epoch": 5.648604269293925,
+      "grad_norm": 1.2661789655685425,
+      "learning_rate": 0.0002,
+      "loss": 0.4264,
+      "step": 5160
+    },
+    {
+      "epoch": 5.659551176792556,
+      "grad_norm": 1.2076870203018188,
+      "learning_rate": 0.0002,
+      "loss": 0.3261,
+      "step": 5170
+    },
+    {
+      "epoch": 5.670498084291188,
+      "grad_norm": 1.2431524991989136,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 5180
+    },
+    {
+      "epoch": 5.681444991789819,
+      "grad_norm": 1.2216639518737793,
+      "learning_rate": 0.0002,
+      "loss": 0.4092,
+      "step": 5190
+    },
+    {
+      "epoch": 5.692391899288451,
+      "grad_norm": 0.9259352684020996,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5200
+    },
+    {
+      "epoch": 5.703338806787083,
+      "grad_norm": 1.7929338216781616,
+      "learning_rate": 0.0002,
+      "loss": 0.3875,
+      "step": 5210
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.4048460721969604,
+      "learning_rate": 0.0002,
+      "loss": 0.4424,
+      "step": 5220
+    },
+    {
+      "epoch": 5.725232621784346,
+      "grad_norm": 1.306874394416809,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 5230
+    },
+    {
+      "epoch": 5.736179529282977,
+      "grad_norm": 1.3137940168380737,
+      "learning_rate": 0.0002,
+      "loss": 0.3889,
+      "step": 5240
+    },
+    {
+      "epoch": 5.747126436781609,
+      "grad_norm": 1.1376476287841797,
+      "learning_rate": 0.0002,
+      "loss": 0.4804,
+      "step": 5250
+    },
+    {
+      "epoch": 5.758073344280241,
+      "grad_norm": 1.450939416885376,
+      "learning_rate": 0.0002,
+      "loss": 0.377,
+      "step": 5260
+    },
+    {
+      "epoch": 5.769020251778873,
+      "grad_norm": 0.983195960521698,
+      "learning_rate": 0.0002,
+      "loss": 0.4732,
+      "step": 5270
+    },
+    {
+      "epoch": 5.779967159277504,
+      "grad_norm": 1.66558837890625,
+      "learning_rate": 0.0002,
+      "loss": 0.4041,
+      "step": 5280
+    },
+    {
+      "epoch": 5.790914066776136,
+      "grad_norm": 0.9789204597473145,
+      "learning_rate": 0.0002,
+      "loss": 0.3643,
+      "step": 5290
+    },
+    {
+      "epoch": 5.801860974274767,
+      "grad_norm": 1.2110556364059448,
+      "learning_rate": 0.0002,
+      "loss": 0.3776,
+      "step": 5300
+    },
+    {
+      "epoch": 5.812807881773399,
+      "grad_norm": 1.3799304962158203,
+      "learning_rate": 0.0002,
+      "loss": 0.4049,
+      "step": 5310
+    },
+    {
+      "epoch": 5.823754789272031,
+      "grad_norm": 1.0570626258850098,
+      "learning_rate": 0.0002,
+      "loss": 0.4362,
+      "step": 5320
+    },
+    {
+      "epoch": 5.834701696770662,
+      "grad_norm": 1.4654436111450195,
+      "learning_rate": 0.0002,
+      "loss": 0.4716,
+      "step": 5330
+    },
+    {
+      "epoch": 5.845648604269294,
+      "grad_norm": 1.5216940641403198,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 5340
+    },
+    {
+      "epoch": 5.856595511767925,
+      "grad_norm": 1.018646001815796,
+      "learning_rate": 0.0002,
+      "loss": 0.3848,
+      "step": 5350
+    },
+    {
+      "epoch": 5.867542419266557,
+      "grad_norm": 1.028951644897461,
+      "learning_rate": 0.0002,
+      "loss": 0.3705,
+      "step": 5360
+    },
+    {
+      "epoch": 5.878489326765189,
+      "grad_norm": 2.571263313293457,
+      "learning_rate": 0.0002,
+      "loss": 0.4213,
+      "step": 5370
+    },
+    {
+      "epoch": 5.889436234263821,
+      "grad_norm": 1.3323984146118164,
+      "learning_rate": 0.0002,
+      "loss": 0.3647,
+      "step": 5380
+    },
+    {
+      "epoch": 5.900383141762452,
+      "grad_norm": 1.4317777156829834,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 5390
+    },
+    {
+      "epoch": 5.911330049261084,
+      "grad_norm": 1.4289140701293945,
+      "learning_rate": 0.0002,
+      "loss": 0.4254,
+      "step": 5400
+    },
+    {
+      "epoch": 5.922276956759715,
+      "grad_norm": 1.3130780458450317,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 5410
+    },
+    {
+      "epoch": 5.933223864258347,
+      "grad_norm": 1.3979902267456055,
+      "learning_rate": 0.0002,
+      "loss": 0.4025,
+      "step": 5420
+    },
+    {
+      "epoch": 5.944170771756979,
+      "grad_norm": 1.1827352046966553,
+      "learning_rate": 0.0002,
+      "loss": 0.3997,
+      "step": 5430
+    },
+    {
+      "epoch": 5.95511767925561,
+      "grad_norm": 1.1672080755233765,
+      "learning_rate": 0.0002,
+      "loss": 0.4163,
+      "step": 5440
+    },
+    {
+      "epoch": 5.966064586754242,
+      "grad_norm": 1.0949620008468628,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 5450
+    },
+    {
+      "epoch": 5.977011494252873,
+      "grad_norm": 1.3183925151824951,
+      "learning_rate": 0.0002,
+      "loss": 0.4219,
+      "step": 5460
+    },
+    {
+      "epoch": 5.987958401751505,
+      "grad_norm": 1.096198320388794,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5470
+    },
+    {
+      "epoch": 5.998905309250137,
+      "grad_norm": 1.2601423263549805,
+      "learning_rate": 0.0002,
+      "loss": 0.3886,
+      "step": 5480
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.611358880996704,
+      "eval_runtime": 46.0638,
+      "eval_samples_per_second": 9.465,
+      "eval_steps_per_second": 1.194,
+      "step": 5481
+    },
+    {
+      "epoch": 6.009852216748769,
+      "grad_norm": 0.9854364991188049,
+      "learning_rate": 0.0002,
+      "loss": 0.2616,
+      "step": 5490
+    },
+    {
+      "epoch": 6.0207991242474,
+      "grad_norm": 1.8073689937591553,
+      "learning_rate": 0.0002,
+      "loss": 0.2412,
+      "step": 5500
+    },
+    {
+      "epoch": 6.031746031746032,
+      "grad_norm": 1.1852164268493652,
+      "learning_rate": 0.0002,
+      "loss": 0.2317,
+      "step": 5510
+    },
+    {
+      "epoch": 6.042692939244663,
+      "grad_norm": 1.0937914848327637,
+      "learning_rate": 0.0002,
+      "loss": 0.224,
+      "step": 5520
+    },
+    {
+      "epoch": 6.053639846743295,
+      "grad_norm": 0.7411194443702698,
+      "learning_rate": 0.0002,
+      "loss": 0.2473,
+      "step": 5530
+    },
+    {
+      "epoch": 6.064586754241927,
+      "grad_norm": 1.552127480506897,
+      "learning_rate": 0.0002,
+      "loss": 0.2846,
+      "step": 5540
+    },
+    {
+      "epoch": 6.075533661740558,
+      "grad_norm": 1.0465604066848755,
+      "learning_rate": 0.0002,
+      "loss": 0.2639,
+      "step": 5550
+    },
+    {
+      "epoch": 6.08648056923919,
+      "grad_norm": 1.4008121490478516,
+      "learning_rate": 0.0002,
+      "loss": 0.2696,
+      "step": 5560
+    },
+    {
+      "epoch": 6.097427476737821,
+      "grad_norm": 1.7049046754837036,
+      "learning_rate": 0.0002,
+      "loss": 0.3049,
+      "step": 5570
+    },
+    {
+      "epoch": 6.108374384236453,
+      "grad_norm": 1.111151933670044,
+      "learning_rate": 0.0002,
+      "loss": 0.263,
+      "step": 5580
+    },
+    {
+      "epoch": 6.119321291735085,
+      "grad_norm": 1.4271087646484375,
+      "learning_rate": 0.0002,
+      "loss": 0.2816,
+      "step": 5590
+    },
+    {
+      "epoch": 6.130268199233717,
+      "grad_norm": 1.3917373418807983,
+      "learning_rate": 0.0002,
+      "loss": 0.2878,
+      "step": 5600
+    },
+    {
+      "epoch": 6.141215106732348,
+      "grad_norm": 1.013689637184143,
+      "learning_rate": 0.0002,
+      "loss": 0.2482,
+      "step": 5610
+    },
+    {
+      "epoch": 6.15216201423098,
+      "grad_norm": 1.342645525932312,
+      "learning_rate": 0.0002,
+      "loss": 0.2841,
+      "step": 5620
+    },
+    {
+      "epoch": 6.163108921729611,
+      "grad_norm": 1.4480562210083008,
+      "learning_rate": 0.0002,
+      "loss": 0.2335,
+      "step": 5630
+    },
+    {
+      "epoch": 6.174055829228243,
+      "grad_norm": 1.2483175992965698,
+      "learning_rate": 0.0002,
+      "loss": 0.2696,
+      "step": 5640
+    },
+    {
+      "epoch": 6.185002736726875,
+      "grad_norm": 1.2944550514221191,
+      "learning_rate": 0.0002,
+      "loss": 0.2656,
+      "step": 5650
+    },
+    {
+      "epoch": 6.195949644225506,
+      "grad_norm": 1.264142632484436,
+      "learning_rate": 0.0002,
+      "loss": 0.2704,
+      "step": 5660
+    },
+    {
+      "epoch": 6.206896551724138,
+      "grad_norm": 1.2068781852722168,
+      "learning_rate": 0.0002,
+      "loss": 0.2971,
+      "step": 5670
+    },
+    {
+      "epoch": 6.217843459222769,
+      "grad_norm": 1.0401629209518433,
+      "learning_rate": 0.0002,
+      "loss": 0.2882,
+      "step": 5680
+    },
+    {
+      "epoch": 6.228790366721401,
+      "grad_norm": 1.2054402828216553,
+      "learning_rate": 0.0002,
+      "loss": 0.3022,
+      "step": 5690
+    },
+    {
+      "epoch": 6.239737274220033,
+      "grad_norm": 1.1278687715530396,
+      "learning_rate": 0.0002,
+      "loss": 0.2949,
+      "step": 5700
+    },
+    {
+      "epoch": 6.250684181718665,
+      "grad_norm": 1.24592125415802,
+      "learning_rate": 0.0002,
+      "loss": 0.2477,
+      "step": 5710
+    },
+    {
+      "epoch": 6.261631089217296,
+      "grad_norm": 1.2686697244644165,
+      "learning_rate": 0.0002,
+      "loss": 0.246,
+      "step": 5720
+    },
+    {
+      "epoch": 6.272577996715928,
+      "grad_norm": 1.1836518049240112,
+      "learning_rate": 0.0002,
+      "loss": 0.2974,
+      "step": 5730
+    },
+    {
+      "epoch": 6.283524904214559,
+      "grad_norm": 1.387752890586853,
+      "learning_rate": 0.0002,
+      "loss": 0.2963,
+      "step": 5740
+    },
+    {
+      "epoch": 6.294471811713191,
+      "grad_norm": 1.9390363693237305,
+      "learning_rate": 0.0002,
+      "loss": 0.2961,
+      "step": 5750
+    },
+    {
+      "epoch": 6.305418719211823,
+      "grad_norm": 1.2919824123382568,
+      "learning_rate": 0.0002,
+      "loss": 0.2765,
+      "step": 5760
+    },
+    {
+      "epoch": 6.316365626710454,
+      "grad_norm": 1.2793965339660645,
+      "learning_rate": 0.0002,
+      "loss": 0.2898,
+      "step": 5770
+    },
+    {
+      "epoch": 6.327312534209086,
+      "grad_norm": 1.5486980676651,
+      "learning_rate": 0.0002,
+      "loss": 0.2786,
+      "step": 5780
+    },
+    {
+      "epoch": 6.338259441707717,
+      "grad_norm": 1.2757408618927002,
+      "learning_rate": 0.0002,
+      "loss": 0.2684,
+      "step": 5790
+    },
+    {
+      "epoch": 6.349206349206349,
+      "grad_norm": 1.3245713710784912,
+      "learning_rate": 0.0002,
+      "loss": 0.2841,
+      "step": 5800
+    },
+    {
+      "epoch": 6.360153256704981,
+      "grad_norm": 1.6262527704238892,
+      "learning_rate": 0.0002,
+      "loss": 0.3096,
+      "step": 5810
+    },
+    {
+      "epoch": 6.371100164203613,
+      "grad_norm": 1.465224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.3219,
+      "step": 5820
+    },
+    {
+      "epoch": 6.382047071702244,
+      "grad_norm": 1.437408447265625,
+      "learning_rate": 0.0002,
+      "loss": 0.2703,
+      "step": 5830
+    },
+    {
+      "epoch": 6.392993979200876,
+      "grad_norm": 1.3094626665115356,
+      "learning_rate": 0.0002,
+      "loss": 0.3012,
+      "step": 5840
+    },
+    {
+      "epoch": 6.403940886699507,
+      "grad_norm": 1.6717544794082642,
+      "learning_rate": 0.0002,
+      "loss": 0.2991,
+      "step": 5850
+    },
+    {
+      "epoch": 6.414887794198139,
+      "grad_norm": 1.1023344993591309,
+      "learning_rate": 0.0002,
+      "loss": 0.2892,
+      "step": 5860
+    },
+    {
+      "epoch": 6.425834701696771,
+      "grad_norm": 1.2397106885910034,
+      "learning_rate": 0.0002,
+      "loss": 0.3078,
+      "step": 5870
+    },
+    {
+      "epoch": 6.436781609195402,
+      "grad_norm": 1.6139185428619385,
+      "learning_rate": 0.0002,
+      "loss": 0.2984,
+      "step": 5880
+    },
+    {
+      "epoch": 6.447728516694034,
+      "grad_norm": 1.3164576292037964,
+      "learning_rate": 0.0002,
+      "loss": 0.2353,
+      "step": 5890
+    },
+    {
+      "epoch": 6.458675424192665,
+      "grad_norm": 1.3317217826843262,
+      "learning_rate": 0.0002,
+      "loss": 0.2772,
+      "step": 5900
+    },
+    {
+      "epoch": 6.469622331691297,
+      "grad_norm": 1.215008020401001,
+      "learning_rate": 0.0002,
+      "loss": 0.2555,
+      "step": 5910
+    },
+    {
+      "epoch": 6.480569239189929,
+      "grad_norm": 1.625672698020935,
+      "learning_rate": 0.0002,
+      "loss": 0.2715,
+      "step": 5920
+    },
+    {
+      "epoch": 6.491516146688561,
+      "grad_norm": 1.1262489557266235,
+      "learning_rate": 0.0002,
+      "loss": 0.2938,
+      "step": 5930
+    },
+    {
+      "epoch": 6.502463054187192,
+      "grad_norm": 1.447100281715393,
+      "learning_rate": 0.0002,
+      "loss": 0.2921,
+      "step": 5940
+    },
+    {
+      "epoch": 6.513409961685824,
+      "grad_norm": 1.3306448459625244,
+      "learning_rate": 0.0002,
+      "loss": 0.3059,
+      "step": 5950
+    },
+    {
+      "epoch": 6.524356869184455,
+      "grad_norm": 1.307732105255127,
+      "learning_rate": 0.0002,
+      "loss": 0.2922,
+      "step": 5960
+    },
+    {
+      "epoch": 6.535303776683087,
+      "grad_norm": 1.1851097345352173,
+      "learning_rate": 0.0002,
+      "loss": 0.2891,
+      "step": 5970
+    },
+    {
+      "epoch": 6.546250684181719,
+      "grad_norm": 1.462816596031189,
+      "learning_rate": 0.0002,
+      "loss": 0.2859,
+      "step": 5980
+    },
+    {
+      "epoch": 6.55719759168035,
+      "grad_norm": 1.2324728965759277,
+      "learning_rate": 0.0002,
+      "loss": 0.2698,
+      "step": 5990
+    },
+    {
+      "epoch": 6.568144499178982,
+      "grad_norm": 1.3627429008483887,
+      "learning_rate": 0.0002,
+      "loss": 0.2672,
+      "step": 6000
+    },
+    {
+      "epoch": 6.579091406677613,
+      "grad_norm": 1.94977867603302,
+      "learning_rate": 0.0002,
+      "loss": 0.3182,
+      "step": 6010
+    },
+    {
+      "epoch": 6.590038314176245,
+      "grad_norm": 1.459844946861267,
+      "learning_rate": 0.0002,
+      "loss": 0.3183,
+      "step": 6020
+    },
+    {
+      "epoch": 6.600985221674877,
+      "grad_norm": 1.4454325437545776,
+      "learning_rate": 0.0002,
+      "loss": 0.3142,
+      "step": 6030
+    },
+    {
+      "epoch": 6.611932129173509,
+      "grad_norm": 1.4245165586471558,
+      "learning_rate": 0.0002,
+      "loss": 0.269,
+      "step": 6040
+    },
+    {
+      "epoch": 6.62287903667214,
+      "grad_norm": 1.195803165435791,
+      "learning_rate": 0.0002,
+      "loss": 0.3041,
+      "step": 6050
+    },
+    {
+      "epoch": 6.633825944170772,
+      "grad_norm": 1.3589898347854614,
+      "learning_rate": 0.0002,
+      "loss": 0.3075,
+      "step": 6060
+    },
+    {
+      "epoch": 6.644772851669403,
+      "grad_norm": 1.3488036394119263,
+      "learning_rate": 0.0002,
+      "loss": 0.3291,
+      "step": 6070
+    },
+    {
+      "epoch": 6.655719759168035,
+      "grad_norm": 1.0954102277755737,
+      "learning_rate": 0.0002,
+      "loss": 0.2898,
+      "step": 6080
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.4431062936782837,
+      "learning_rate": 0.0002,
+      "loss": 0.3489,
+      "step": 6090
+    },
+    {
+      "epoch": 6.677613574165298,
+      "grad_norm": 1.4387465715408325,
+      "learning_rate": 0.0002,
+      "loss": 0.2816,
+      "step": 6100
+    },
+    {
+      "epoch": 6.68856048166393,
+      "grad_norm": 1.8398990631103516,
+      "learning_rate": 0.0002,
+      "loss": 0.2705,
+      "step": 6110
+    },
+    {
+      "epoch": 6.699507389162561,
+      "grad_norm": 1.3523273468017578,
+      "learning_rate": 0.0002,
+      "loss": 0.3214,
+      "step": 6120
+    },
+    {
+      "epoch": 6.710454296661193,
+      "grad_norm": 1.6326191425323486,
+      "learning_rate": 0.0002,
+      "loss": 0.287,
+      "step": 6130
+    },
+    {
+      "epoch": 6.721401204159825,
+      "grad_norm": 1.3677960634231567,
+      "learning_rate": 0.0002,
+      "loss": 0.2857,
+      "step": 6140
+    },
+    {
+      "epoch": 6.732348111658457,
+      "grad_norm": 1.1993201971054077,
+      "learning_rate": 0.0002,
+      "loss": 0.3264,
+      "step": 6150
+    },
+    {
+      "epoch": 6.743295019157088,
+      "grad_norm": 1.1864078044891357,
+      "learning_rate": 0.0002,
+      "loss": 0.3071,
+      "step": 6160
+    },
+    {
+      "epoch": 6.75424192665572,
+      "grad_norm": 1.1625522375106812,
+      "learning_rate": 0.0002,
+      "loss": 0.3087,
+      "step": 6170
+    },
+    {
+      "epoch": 6.765188834154351,
+      "grad_norm": 1.5803234577178955,
+      "learning_rate": 0.0002,
+      "loss": 0.3551,
+      "step": 6180
+    },
+    {
+      "epoch": 6.776135741652983,
+      "grad_norm": 1.151746153831482,
+      "learning_rate": 0.0002,
+      "loss": 0.3059,
+      "step": 6190
+    },
+    {
+      "epoch": 6.787082649151615,
+      "grad_norm": 1.0727161169052124,
+      "learning_rate": 0.0002,
+      "loss": 0.2697,
+      "step": 6200
+    },
+    {
+      "epoch": 6.798029556650246,
+      "grad_norm": 1.4148162603378296,
+      "learning_rate": 0.0002,
+      "loss": 0.2844,
+      "step": 6210
+    },
+    {
+      "epoch": 6.808976464148878,
+      "grad_norm": 1.2071447372436523,
+      "learning_rate": 0.0002,
+      "loss": 0.3417,
+      "step": 6220
+    },
+    {
+      "epoch": 6.819923371647509,
+      "grad_norm": 1.3843804597854614,
+      "learning_rate": 0.0002,
+      "loss": 0.3066,
+      "step": 6230
+    },
+    {
+      "epoch": 6.830870279146141,
+      "grad_norm": 1.2490662336349487,
+      "learning_rate": 0.0002,
+      "loss": 0.2769,
+      "step": 6240
+    },
+    {
+      "epoch": 6.841817186644773,
+      "grad_norm": 1.6029689311981201,
+      "learning_rate": 0.0002,
+      "loss": 0.3237,
+      "step": 6250
+    },
+    {
+      "epoch": 6.852764094143405,
+      "grad_norm": 1.0388455390930176,
+      "learning_rate": 0.0002,
+      "loss": 0.3152,
+      "step": 6260
+    },
+    {
+      "epoch": 6.863711001642036,
+      "grad_norm": 1.3883857727050781,
+      "learning_rate": 0.0002,
+      "loss": 0.3026,
+      "step": 6270
+    },
+    {
+      "epoch": 6.874657909140668,
+      "grad_norm": 1.0500187873840332,
+      "learning_rate": 0.0002,
+      "loss": 0.3175,
+      "step": 6280
+    },
+    {
+      "epoch": 6.885604816639299,
+      "grad_norm": 1.4243487119674683,
+      "learning_rate": 0.0002,
+      "loss": 0.2952,
+      "step": 6290
+    },
+    {
+      "epoch": 6.896551724137931,
+      "grad_norm": 1.3169665336608887,
+      "learning_rate": 0.0002,
+      "loss": 0.2679,
+      "step": 6300
+    },
+    {
+      "epoch": 6.907498631636563,
+      "grad_norm": 1.5261493921279907,
+      "learning_rate": 0.0002,
+      "loss": 0.3291,
+      "step": 6310
+    },
+    {
+      "epoch": 6.9184455391351944,
+      "grad_norm": 1.578403115272522,
+      "learning_rate": 0.0002,
+      "loss": 0.3344,
+      "step": 6320
+    },
+    {
+      "epoch": 6.929392446633826,
+      "grad_norm": 1.4093263149261475,
+      "learning_rate": 0.0002,
+      "loss": 0.3263,
+      "step": 6330
+    },
+    {
+      "epoch": 6.940339354132457,
+      "grad_norm": 1.4003552198410034,
+      "learning_rate": 0.0002,
+      "loss": 0.3396,
+      "step": 6340
+    },
+    {
+      "epoch": 6.951286261631089,
+      "grad_norm": 1.650190830230713,
+      "learning_rate": 0.0002,
+      "loss": 0.3476,
+      "step": 6350
+    },
+    {
+      "epoch": 6.962233169129721,
+      "grad_norm": 1.2314515113830566,
+      "learning_rate": 0.0002,
+      "loss": 0.3442,
+      "step": 6360
+    },
+    {
+      "epoch": 6.973180076628353,
+      "grad_norm": 1.270980954170227,
+      "learning_rate": 0.0002,
+      "loss": 0.3341,
+      "step": 6370
+    },
+    {
+      "epoch": 6.984126984126984,
+      "grad_norm": 1.6352545022964478,
+      "learning_rate": 0.0002,
+      "loss": 0.3425,
+      "step": 6380
+    },
+    {
+      "epoch": 6.995073891625616,
+      "grad_norm": 1.3744925260543823,
+      "learning_rate": 0.0002,
+      "loss": 0.3647,
+      "step": 6390
+    },
+    {
+      "epoch": 6.999452654625069,
+      "eval_loss": 1.756764531135559,
+      "eval_runtime": 46.0542,
+      "eval_samples_per_second": 9.467,
+      "eval_steps_per_second": 1.194,
+      "step": 6394
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.284454714192691e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-6394/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ec5568997301ef15ec19890e1db44076874b660d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e75d7b31a1782d17b5e60ef83b5a281f3a63ddf3550ecdd8d670c015ab44a3c3
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..36409135a3bc94b0a9f063bc966f43a9cb962cbd
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c634d73066f5266d0914318b725cc5ae40d7bb1c13681976f5672d9895120ca
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..cbed656ccd7cafa4f68547e8cd86492c6d0ef2f4
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5a84bd578fc3f9ab30db5f881ce38f2483cd31a0966710b640743309d399fd
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3f0b65f305d95bfee75fc360c1debb95cf87eb45
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8de44f7d4d23605f9cddf0b9002140a8423325f3c130a3fe1b1f6033c2d9861e
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..dc82a795067083bef62bc9d23a6705187ed03203
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/trainer_state.json
@@ -0,0 +1,5207 @@
+{
+  "best_metric": 1.14472496509552,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827",
+  "epoch": 7.995621237000547,
+  "eval_steps": 10,
+  "global_step": 7304,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    },
+    {
+      "epoch": 1.0071154898741106,
+      "grad_norm": 0.37776654958724976,
+      "learning_rate": 0.0002,
+      "loss": 1.0322,
+      "step": 920
+    },
+    {
+      "epoch": 1.0180623973727423,
+      "grad_norm": 0.31784629821777344,
+      "learning_rate": 0.0002,
+      "loss": 1.1346,
+      "step": 930
+    },
+    {
+      "epoch": 1.0290093048713738,
+      "grad_norm": 0.24244336783885956,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 940
+    },
+    {
+      "epoch": 1.0399562123700055,
+      "grad_norm": 0.3185454308986664,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 950
+    },
+    {
+      "epoch": 1.0509031198686372,
+      "grad_norm": 0.3589441478252411,
+      "learning_rate": 0.0002,
+      "loss": 1.1645,
+      "step": 960
+    },
+    {
+      "epoch": 1.0618500273672686,
+      "grad_norm": 0.38593578338623047,
+      "learning_rate": 0.0002,
+      "loss": 1.0317,
+      "step": 970
+    },
+    {
+      "epoch": 1.0727969348659003,
+      "grad_norm": 0.39694955945014954,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 980
+    },
+    {
+      "epoch": 1.083743842364532,
+      "grad_norm": 0.469817191362381,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 990
+    },
+    {
+      "epoch": 1.0946907498631637,
+      "grad_norm": 0.2634755074977875,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1056376573617952,
+      "grad_norm": 0.43189436197280884,
+      "learning_rate": 0.0002,
+      "loss": 1.0144,
+      "step": 1010
+    },
+    {
+      "epoch": 1.116584564860427,
+      "grad_norm": 0.5559977889060974,
+      "learning_rate": 0.0002,
+      "loss": 0.9663,
+      "step": 1020
+    },
+    {
+      "epoch": 1.1275314723590586,
+      "grad_norm": 0.32100191712379456,
+      "learning_rate": 0.0002,
+      "loss": 1.0481,
+      "step": 1030
+    },
+    {
+      "epoch": 1.1384783798576903,
+      "grad_norm": 0.40179768204689026,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 1040
+    },
+    {
+      "epoch": 1.1494252873563218,
+      "grad_norm": 0.3659493029117584,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 1050
+    },
+    {
+      "epoch": 1.1603721948549535,
+      "grad_norm": 0.701704204082489,
+      "learning_rate": 0.0002,
+      "loss": 0.9597,
+      "step": 1060
+    },
+    {
+      "epoch": 1.1713191023535852,
+      "grad_norm": 0.3650563359260559,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1822660098522166,
+      "grad_norm": 0.3191976249217987,
+      "learning_rate": 0.0002,
+      "loss": 0.907,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1932129173508483,
+      "grad_norm": 0.3615441918373108,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 1090
+    },
+    {
+      "epoch": 1.20415982484948,
+      "grad_norm": 0.39474231004714966,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2151067323481117,
+      "grad_norm": 0.3752822279930115,
+      "learning_rate": 0.0002,
+      "loss": 0.9852,
+      "step": 1110
+    },
+    {
+      "epoch": 1.2260536398467432,
+      "grad_norm": 0.4165991246700287,
+      "learning_rate": 0.0002,
+      "loss": 1.0067,
+      "step": 1120
+    },
+    {
+      "epoch": 1.237000547345375,
+      "grad_norm": 0.5326506495475769,
+      "learning_rate": 0.0002,
+      "loss": 1.0244,
+      "step": 1130
+    },
+    {
+      "epoch": 1.2479474548440066,
+      "grad_norm": 0.48845794796943665,
+      "learning_rate": 0.0002,
+      "loss": 1.0542,
+      "step": 1140
+    },
+    {
+      "epoch": 1.2588943623426383,
+      "grad_norm": 0.29910150170326233,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 1150
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 0.5069725513458252,
+      "learning_rate": 0.0002,
+      "loss": 1.2233,
+      "step": 1160
+    },
+    {
+      "epoch": 1.2807881773399015,
+      "grad_norm": 0.29500406980514526,
+      "learning_rate": 0.0002,
+      "loss": 1.0992,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2917350848385332,
+      "grad_norm": 0.4711538851261139,
+      "learning_rate": 0.0002,
+      "loss": 1.0291,
+      "step": 1180
+    },
+    {
+      "epoch": 1.3026819923371646,
+      "grad_norm": 0.4203340709209442,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1190
+    },
+    {
+      "epoch": 1.3136288998357963,
+      "grad_norm": 0.36101874709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1200
+    },
+    {
+      "epoch": 1.324575807334428,
+      "grad_norm": 0.4608800411224365,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 1210
+    },
+    {
+      "epoch": 1.3355227148330597,
+      "grad_norm": 0.6570906639099121,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 1220
+    },
+    {
+      "epoch": 1.3464696223316914,
+      "grad_norm": 0.5352164506912231,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 1230
+    },
+    {
+      "epoch": 1.357416529830323,
+      "grad_norm": 0.3885001242160797,
+      "learning_rate": 0.0002,
+      "loss": 1.0513,
+      "step": 1240
+    },
+    {
+      "epoch": 1.3683634373289546,
+      "grad_norm": 0.2987913489341736,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 1250
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.42070427536964417,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 1260
+    },
+    {
+      "epoch": 1.3902572523262178,
+      "grad_norm": 0.5957782864570618,
+      "learning_rate": 0.0002,
+      "loss": 1.1092,
+      "step": 1270
+    },
+    {
+      "epoch": 1.4012041598248495,
+      "grad_norm": 0.32898882031440735,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 1280
+    },
+    {
+      "epoch": 1.4121510673234812,
+      "grad_norm": 0.27624452114105225,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 1290
+    },
+    {
+      "epoch": 1.4230979748221126,
+      "grad_norm": 0.49570828676223755,
+      "learning_rate": 0.0002,
+      "loss": 1.024,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4340448823207443,
+      "grad_norm": 0.26191383600234985,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 1310
+    },
+    {
+      "epoch": 1.444991789819376,
+      "grad_norm": 0.35664042830467224,
+      "learning_rate": 0.0002,
+      "loss": 1.0788,
+      "step": 1320
+    },
+    {
+      "epoch": 1.4559386973180077,
+      "grad_norm": 0.45126354694366455,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 1330
+    },
+    {
+      "epoch": 1.4668856048166394,
+      "grad_norm": 0.37318357825279236,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 1340
+    },
+    {
+      "epoch": 1.477832512315271,
+      "grad_norm": 0.6428970098495483,
+      "learning_rate": 0.0002,
+      "loss": 1.0507,
+      "step": 1350
+    },
+    {
+      "epoch": 1.4887794198139026,
+      "grad_norm": 0.43256187438964844,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 1360
+    },
+    {
+      "epoch": 1.4997263273125343,
+      "grad_norm": 0.5343793630599976,
+      "learning_rate": 0.0002,
+      "loss": 1.1321,
+      "step": 1370
+    },
+    {
+      "epoch": 1.5106732348111658,
+      "grad_norm": 0.315437376499176,
+      "learning_rate": 0.0002,
+      "loss": 1.1054,
+      "step": 1380
+    },
+    {
+      "epoch": 1.5216201423097975,
+      "grad_norm": 0.41561153531074524,
+      "learning_rate": 0.0002,
+      "loss": 0.8916,
+      "step": 1390
+    },
+    {
+      "epoch": 1.5325670498084292,
+      "grad_norm": 0.3201070725917816,
+      "learning_rate": 0.0002,
+      "loss": 1.0391,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5435139573070606,
+      "grad_norm": 0.505537211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 1410
+    },
+    {
+      "epoch": 1.5544608648056923,
+      "grad_norm": 0.3747410178184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0775,
+      "step": 1420
+    },
+    {
+      "epoch": 1.565407772304324,
+      "grad_norm": 0.49385908246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1430
+    },
+    {
+      "epoch": 1.5763546798029555,
+      "grad_norm": 0.49831628799438477,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 1440
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 0.372127890586853,
+      "learning_rate": 0.0002,
+      "loss": 1.0079,
+      "step": 1450
+    },
+    {
+      "epoch": 1.598248494800219,
+      "grad_norm": 0.40070840716362,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 1460
+    },
+    {
+      "epoch": 1.6091954022988506,
+      "grad_norm": 0.34907400608062744,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 1470
+    },
+    {
+      "epoch": 1.6201423097974823,
+      "grad_norm": 0.4632418751716614,
+      "learning_rate": 0.0002,
+      "loss": 0.9743,
+      "step": 1480
+    },
+    {
+      "epoch": 1.6310892172961138,
+      "grad_norm": 0.40164515376091003,
+      "learning_rate": 0.0002,
+      "loss": 1.0103,
+      "step": 1490
+    },
+    {
+      "epoch": 1.6420361247947455,
+      "grad_norm": 0.3214994966983795,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6529830322933772,
+      "grad_norm": 0.3727897107601166,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 1510
+    },
+    {
+      "epoch": 1.6639299397920086,
+      "grad_norm": 0.3817640542984009,
+      "learning_rate": 0.0002,
+      "loss": 1.0443,
+      "step": 1520
+    },
+    {
+      "epoch": 1.6748768472906403,
+      "grad_norm": 0.5592136979103088,
+      "learning_rate": 0.0002,
+      "loss": 1.0511,
+      "step": 1530
+    },
+    {
+      "epoch": 1.685823754789272,
+      "grad_norm": 0.44636598229408264,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 1540
+    },
+    {
+      "epoch": 1.6967706622879035,
+      "grad_norm": 0.40441709756851196,
+      "learning_rate": 0.0002,
+      "loss": 1.033,
+      "step": 1550
+    },
+    {
+      "epoch": 1.7077175697865354,
+      "grad_norm": 0.3243522644042969,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1560
+    },
+    {
+      "epoch": 1.718664477285167,
+      "grad_norm": 0.34277570247650146,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1570
+    },
+    {
+      "epoch": 1.7296113847837986,
+      "grad_norm": 0.3279995024204254,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 1580
+    },
+    {
+      "epoch": 1.7405582922824303,
+      "grad_norm": 0.41968777775764465,
+      "learning_rate": 0.0002,
+      "loss": 0.9414,
+      "step": 1590
+    },
+    {
+      "epoch": 1.7515051997810618,
+      "grad_norm": 0.39464613795280457,
+      "learning_rate": 0.0002,
+      "loss": 0.9768,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7624521072796935,
+      "grad_norm": 0.3839009404182434,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 1610
+    },
+    {
+      "epoch": 1.7733990147783252,
+      "grad_norm": 0.3250715434551239,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 1620
+    },
+    {
+      "epoch": 1.7843459222769567,
+      "grad_norm": 0.5166561007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.0049,
+      "step": 1630
+    },
+    {
+      "epoch": 1.7952928297755884,
+      "grad_norm": 0.4115183353424072,
+      "learning_rate": 0.0002,
+      "loss": 1.0364,
+      "step": 1640
+    },
+    {
+      "epoch": 1.80623973727422,
+      "grad_norm": 0.373780220746994,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 1650
+    },
+    {
+      "epoch": 1.8171866447728515,
+      "grad_norm": 0.49697014689445496,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 1660
+    },
+    {
+      "epoch": 1.8281335522714834,
+      "grad_norm": 1.0308938026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 1670
+    },
+    {
+      "epoch": 1.839080459770115,
+      "grad_norm": 0.4851366877555847,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 1680
+    },
+    {
+      "epoch": 1.8500273672687466,
+      "grad_norm": 0.3262481391429901,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 1690
+    },
+    {
+      "epoch": 1.8609742747673783,
+      "grad_norm": 0.6904496550559998,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8719211822660098,
+      "grad_norm": 0.49789851903915405,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 1710
+    },
+    {
+      "epoch": 1.8828680897646415,
+      "grad_norm": 0.3035794198513031,
+      "learning_rate": 0.0002,
+      "loss": 1.0258,
+      "step": 1720
+    },
+    {
+      "epoch": 1.8938149972632732,
+      "grad_norm": 0.4588414430618286,
+      "learning_rate": 0.0002,
+      "loss": 0.9916,
+      "step": 1730
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.4313034117221832,
+      "learning_rate": 0.0002,
+      "loss": 0.9526,
+      "step": 1740
+    },
+    {
+      "epoch": 1.9157088122605364,
+      "grad_norm": 0.38562044501304626,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 1750
+    },
+    {
+      "epoch": 1.926655719759168,
+      "grad_norm": 0.46947410702705383,
+      "learning_rate": 0.0002,
+      "loss": 0.9763,
+      "step": 1760
+    },
+    {
+      "epoch": 1.9376026272577995,
+      "grad_norm": 0.3848404884338379,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 1770
+    },
+    {
+      "epoch": 1.9485495347564314,
+      "grad_norm": 0.30422744154930115,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 1780
+    },
+    {
+      "epoch": 1.959496442255063,
+      "grad_norm": 0.41100990772247314,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 1790
+    },
+    {
+      "epoch": 1.9704433497536946,
+      "grad_norm": 0.3492335081100464,
+      "learning_rate": 0.0002,
+      "loss": 1.003,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9813902572523263,
+      "grad_norm": 0.364577978849411,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 1810
+    },
+    {
+      "epoch": 1.9923371647509578,
+      "grad_norm": 0.4312075674533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0699,
+      "step": 1820
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.14472496509552,
+      "eval_runtime": 46.0786,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 1827
+    },
+    {
+      "epoch": 2.0032840722495897,
+      "grad_norm": 0.5989689230918884,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 1830
+    },
+    {
+      "epoch": 2.014230979748221,
+      "grad_norm": 0.49720922112464905,
+      "learning_rate": 0.0002,
+      "loss": 0.9384,
+      "step": 1840
+    },
+    {
+      "epoch": 2.0251778872468527,
+      "grad_norm": 0.42675456404685974,
+      "learning_rate": 0.0002,
+      "loss": 0.8482,
+      "step": 1850
+    },
+    {
+      "epoch": 2.0361247947454846,
+      "grad_norm": 0.4637208580970764,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 1860
+    },
+    {
+      "epoch": 2.047071702244116,
+      "grad_norm": 0.8329976797103882,
+      "learning_rate": 0.0002,
+      "loss": 0.8212,
+      "step": 1870
+    },
+    {
+      "epoch": 2.0580186097427475,
+      "grad_norm": 0.7869427800178528,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 1880
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.4927455186843872,
+      "learning_rate": 0.0002,
+      "loss": 0.9351,
+      "step": 1890
+    },
+    {
+      "epoch": 2.079912424740011,
+      "grad_norm": 0.6264246702194214,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0908593322386424,
+      "grad_norm": 1.1164122819900513,
+      "learning_rate": 0.0002,
+      "loss": 0.9095,
+      "step": 1910
+    },
+    {
+      "epoch": 2.1018062397372743,
+      "grad_norm": 0.5283981561660767,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 1920
+    },
+    {
+      "epoch": 2.112753147235906,
+      "grad_norm": 0.45621731877326965,
+      "learning_rate": 0.0002,
+      "loss": 0.8065,
+      "step": 1930
+    },
+    {
+      "epoch": 2.1237000547345373,
+      "grad_norm": 1.381791591644287,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 1940
+    },
+    {
+      "epoch": 2.134646962233169,
+      "grad_norm": 0.5151259899139404,
+      "learning_rate": 0.0002,
+      "loss": 0.9006,
+      "step": 1950
+    },
+    {
+      "epoch": 2.1455938697318007,
+      "grad_norm": 0.9806339740753174,
+      "learning_rate": 0.0002,
+      "loss": 0.8436,
+      "step": 1960
+    },
+    {
+      "epoch": 2.1565407772304326,
+      "grad_norm": 0.4734154939651489,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 1970
+    },
+    {
+      "epoch": 2.167487684729064,
+      "grad_norm": 0.9553168416023254,
+      "learning_rate": 0.0002,
+      "loss": 0.9172,
+      "step": 1980
+    },
+    {
+      "epoch": 2.1784345922276955,
+      "grad_norm": 0.5895838141441345,
+      "learning_rate": 0.0002,
+      "loss": 0.8047,
+      "step": 1990
+    },
+    {
+      "epoch": 2.1893814997263275,
+      "grad_norm": 0.4488855302333832,
+      "learning_rate": 0.0002,
+      "loss": 0.7841,
+      "step": 2000
+    },
+    {
+      "epoch": 2.200328407224959,
+      "grad_norm": 1.0760235786437988,
+      "learning_rate": 0.0002,
+      "loss": 0.8205,
+      "step": 2010
+    },
+    {
+      "epoch": 2.2112753147235904,
+      "grad_norm": 0.5038785338401794,
+      "learning_rate": 0.0002,
+      "loss": 0.7923,
+      "step": 2020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.59819495677948,
+      "learning_rate": 0.0002,
+      "loss": 0.8973,
+      "step": 2030
+    },
+    {
+      "epoch": 2.233169129720854,
+      "grad_norm": 0.5012075304985046,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 2040
+    },
+    {
+      "epoch": 2.2441160372194853,
+      "grad_norm": 0.44978439807891846,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 2050
+    },
+    {
+      "epoch": 2.255062944718117,
+      "grad_norm": 0.5350462198257446,
+      "learning_rate": 0.0002,
+      "loss": 0.823,
+      "step": 2060
+    },
+    {
+      "epoch": 2.2660098522167487,
+      "grad_norm": 0.6020669937133789,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 2070
+    },
+    {
+      "epoch": 2.2769567597153806,
+      "grad_norm": 0.5246821045875549,
+      "learning_rate": 0.0002,
+      "loss": 0.8023,
+      "step": 2080
+    },
+    {
+      "epoch": 2.287903667214012,
+      "grad_norm": 0.5711268782615662,
+      "learning_rate": 0.0002,
+      "loss": 0.8984,
+      "step": 2090
+    },
+    {
+      "epoch": 2.2988505747126435,
+      "grad_norm": 0.617317259311676,
+      "learning_rate": 0.0002,
+      "loss": 0.9093,
+      "step": 2100
+    },
+    {
+      "epoch": 2.3097974822112755,
+      "grad_norm": 0.8608947396278381,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2110
+    },
+    {
+      "epoch": 2.320744389709907,
+      "grad_norm": 0.4739076793193817,
+      "learning_rate": 0.0002,
+      "loss": 0.7839,
+      "step": 2120
+    },
+    {
+      "epoch": 2.3316912972085384,
+      "grad_norm": 0.5538856983184814,
+      "learning_rate": 0.0002,
+      "loss": 0.84,
+      "step": 2130
+    },
+    {
+      "epoch": 2.3426382047071703,
+      "grad_norm": 0.6064935326576233,
+      "learning_rate": 0.0002,
+      "loss": 0.8994,
+      "step": 2140
+    },
+    {
+      "epoch": 2.353585112205802,
+      "grad_norm": 0.5019068121910095,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 2150
+    },
+    {
+      "epoch": 2.3645320197044333,
+      "grad_norm": 0.45340514183044434,
+      "learning_rate": 0.0002,
+      "loss": 0.9576,
+      "step": 2160
+    },
+    {
+      "epoch": 2.375478927203065,
+      "grad_norm": 0.7347203493118286,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 2170
+    },
+    {
+      "epoch": 2.3864258347016967,
+      "grad_norm": 0.46922534704208374,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 2180
+    },
+    {
+      "epoch": 2.3973727422003286,
+      "grad_norm": 0.5507845878601074,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2190
+    },
+    {
+      "epoch": 2.40831964969896,
+      "grad_norm": 0.5621911883354187,
+      "learning_rate": 0.0002,
+      "loss": 0.8178,
+      "step": 2200
+    },
+    {
+      "epoch": 2.4192665571975915,
+      "grad_norm": 0.5023514032363892,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 2210
+    },
+    {
+      "epoch": 2.4302134646962235,
+      "grad_norm": 0.6124861240386963,
+      "learning_rate": 0.0002,
+      "loss": 0.8355,
+      "step": 2220
+    },
+    {
+      "epoch": 2.441160372194855,
+      "grad_norm": 0.49614205956459045,
+      "learning_rate": 0.0002,
+      "loss": 0.885,
+      "step": 2230
+    },
+    {
+      "epoch": 2.4521072796934864,
+      "grad_norm": 0.6477900743484497,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 2240
+    },
+    {
+      "epoch": 2.4630541871921183,
+      "grad_norm": 0.5868843793869019,
+      "learning_rate": 0.0002,
+      "loss": 0.8622,
+      "step": 2250
+    },
+    {
+      "epoch": 2.47400109469075,
+      "grad_norm": 0.4364610016345978,
+      "learning_rate": 0.0002,
+      "loss": 0.8498,
+      "step": 2260
+    },
+    {
+      "epoch": 2.4849480021893813,
+      "grad_norm": 0.5792964696884155,
+      "learning_rate": 0.0002,
+      "loss": 0.8378,
+      "step": 2270
+    },
+    {
+      "epoch": 2.495894909688013,
+      "grad_norm": 0.5421269536018372,
+      "learning_rate": 0.0002,
+      "loss": 0.8743,
+      "step": 2280
+    },
+    {
+      "epoch": 2.5068418171866447,
+      "grad_norm": 0.5525493025779724,
+      "learning_rate": 0.0002,
+      "loss": 0.9637,
+      "step": 2290
+    },
+    {
+      "epoch": 2.5177887246852766,
+      "grad_norm": 0.6463850140571594,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 2300
+    },
+    {
+      "epoch": 2.528735632183908,
+      "grad_norm": 0.6861311793327332,
+      "learning_rate": 0.0002,
+      "loss": 0.8591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 0.5563555359840393,
+      "learning_rate": 0.0002,
+      "loss": 0.9287,
+      "step": 2320
+    },
+    {
+      "epoch": 2.5506294471811715,
+      "grad_norm": 0.5721169114112854,
+      "learning_rate": 0.0002,
+      "loss": 0.945,
+      "step": 2330
+    },
+    {
+      "epoch": 2.561576354679803,
+      "grad_norm": 0.5258274674415588,
+      "learning_rate": 0.0002,
+      "loss": 0.8271,
+      "step": 2340
+    },
+    {
+      "epoch": 2.572523262178435,
+      "grad_norm": 0.7057380676269531,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 2350
+    },
+    {
+      "epoch": 2.5834701696770663,
+      "grad_norm": 0.6869027614593506,
+      "learning_rate": 0.0002,
+      "loss": 0.8615,
+      "step": 2360
+    },
+    {
+      "epoch": 2.594417077175698,
+      "grad_norm": 0.4960809648036957,
+      "learning_rate": 0.0002,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 2.6053639846743293,
+      "grad_norm": 0.9288380146026611,
+      "learning_rate": 0.0002,
+      "loss": 0.8476,
+      "step": 2380
+    },
+    {
+      "epoch": 2.616310892172961,
+      "grad_norm": 0.3765334188938141,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 2390
+    },
+    {
+      "epoch": 2.6272577996715927,
+      "grad_norm": 0.7487865686416626,
+      "learning_rate": 0.0002,
+      "loss": 0.8764,
+      "step": 2400
+    },
+    {
+      "epoch": 2.6382047071702246,
+      "grad_norm": 0.6141156554222107,
+      "learning_rate": 0.0002,
+      "loss": 0.7577,
+      "step": 2410
+    },
+    {
+      "epoch": 2.649151614668856,
+      "grad_norm": 0.8420507907867432,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 2.6600985221674875,
+      "grad_norm": 0.53386390209198,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 2430
+    },
+    {
+      "epoch": 2.6710454296661195,
+      "grad_norm": 0.5520607233047485,
+      "learning_rate": 0.0002,
+      "loss": 0.8486,
+      "step": 2440
+    },
+    {
+      "epoch": 2.681992337164751,
+      "grad_norm": 0.5337599515914917,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 2450
+    },
+    {
+      "epoch": 2.692939244663383,
+      "grad_norm": 0.48790836334228516,
+      "learning_rate": 0.0002,
+      "loss": 0.792,
+      "step": 2460
+    },
+    {
+      "epoch": 2.7038861521620143,
+      "grad_norm": 0.8287786245346069,
+      "learning_rate": 0.0002,
+      "loss": 0.7864,
+      "step": 2470
+    },
+    {
+      "epoch": 2.714833059660646,
+      "grad_norm": 0.5876168608665466,
+      "learning_rate": 0.0002,
+      "loss": 0.8244,
+      "step": 2480
+    },
+    {
+      "epoch": 2.7257799671592773,
+      "grad_norm": 0.5206760764122009,
+      "learning_rate": 0.0002,
+      "loss": 0.8773,
+      "step": 2490
+    },
+    {
+      "epoch": 2.736726874657909,
+      "grad_norm": 0.5619136691093445,
+      "learning_rate": 0.0002,
+      "loss": 0.8097,
+      "step": 2500
+    },
+    {
+      "epoch": 2.7476737821565407,
+      "grad_norm": 0.5614883899688721,
+      "learning_rate": 0.0002,
+      "loss": 0.8377,
+      "step": 2510
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.6157700419425964,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2520
+    },
+    {
+      "epoch": 2.769567597153804,
+      "grad_norm": 0.5529953837394714,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 2530
+    },
+    {
+      "epoch": 2.7805145046524355,
+      "grad_norm": 0.6731224060058594,
+      "learning_rate": 0.0002,
+      "loss": 0.8881,
+      "step": 2540
+    },
+    {
+      "epoch": 2.7914614121510675,
+      "grad_norm": 0.6960386633872986,
+      "learning_rate": 0.0002,
+      "loss": 0.9125,
+      "step": 2550
+    },
+    {
+      "epoch": 2.802408319649699,
+      "grad_norm": 0.5203493237495422,
+      "learning_rate": 0.0002,
+      "loss": 0.7823,
+      "step": 2560
+    },
+    {
+      "epoch": 2.813355227148331,
+      "grad_norm": 1.036837100982666,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 2570
+    },
+    {
+      "epoch": 2.8243021346469623,
+      "grad_norm": 0.6125805377960205,
+      "learning_rate": 0.0002,
+      "loss": 0.8427,
+      "step": 2580
+    },
+    {
+      "epoch": 2.835249042145594,
+      "grad_norm": 0.6298092603683472,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 2590
+    },
+    {
+      "epoch": 2.8461959496442253,
+      "grad_norm": 0.5882203578948975,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2600
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.8619399666786194,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 2610
+    },
+    {
+      "epoch": 2.8680897646414887,
+      "grad_norm": 0.4722687304019928,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 2620
+    },
+    {
+      "epoch": 2.8790366721401206,
+      "grad_norm": 0.47399574518203735,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 2630
+    },
+    {
+      "epoch": 2.889983579638752,
+      "grad_norm": 0.5639172792434692,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2640
+    },
+    {
+      "epoch": 2.9009304871373836,
+      "grad_norm": 0.4676816761493683,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 2650
+    },
+    {
+      "epoch": 2.9118773946360155,
+      "grad_norm": 0.6906291246414185,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2660
+    },
+    {
+      "epoch": 2.922824302134647,
+      "grad_norm": 0.4369746148586273,
+      "learning_rate": 0.0002,
+      "loss": 0.8966,
+      "step": 2670
+    },
+    {
+      "epoch": 2.933771209633279,
+      "grad_norm": 0.46423083543777466,
+      "learning_rate": 0.0002,
+      "loss": 0.9173,
+      "step": 2680
+    },
+    {
+      "epoch": 2.9447181171319103,
+      "grad_norm": 0.5700525045394897,
+      "learning_rate": 0.0002,
+      "loss": 0.7961,
+      "step": 2690
+    },
+    {
+      "epoch": 2.955665024630542,
+      "grad_norm": 0.6221476793289185,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 2700
+    },
+    {
+      "epoch": 2.9666119321291733,
+      "grad_norm": 0.6102682948112488,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 2710
+    },
+    {
+      "epoch": 2.977558839627805,
+      "grad_norm": 0.5317878723144531,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2720
+    },
+    {
+      "epoch": 2.9885057471264367,
+      "grad_norm": 0.4438510835170746,
+      "learning_rate": 0.0002,
+      "loss": 0.9314,
+      "step": 2730
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "grad_norm": 0.5022130012512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 2740
+    },
+    {
+      "epoch": 2.9994526546250686,
+      "eval_loss": 1.1722838878631592,
+      "eval_runtime": 46.0829,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.194,
+      "step": 2740
+    },
+    {
+      "epoch": 3.0103995621237,
+      "grad_norm": 0.6384502053260803,
+      "learning_rate": 0.0002,
+      "loss": 0.6443,
+      "step": 2750
+    },
+    {
+      "epoch": 3.0213464696223316,
+      "grad_norm": 0.9928722381591797,
+      "learning_rate": 0.0002,
+      "loss": 0.7123,
+      "step": 2760
+    },
+    {
+      "epoch": 3.0322933771209635,
+      "grad_norm": 0.7813051342964172,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 2770
+    },
+    {
+      "epoch": 3.043240284619595,
+      "grad_norm": 1.0202556848526,
+      "learning_rate": 0.0002,
+      "loss": 0.6042,
+      "step": 2780
+    },
+    {
+      "epoch": 3.0541871921182264,
+      "grad_norm": 0.7581062316894531,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 2790
+    },
+    {
+      "epoch": 3.0651340996168583,
+      "grad_norm": 0.6252710223197937,
+      "learning_rate": 0.0002,
+      "loss": 0.6349,
+      "step": 2800
+    },
+    {
+      "epoch": 3.07608100711549,
+      "grad_norm": 0.7738662958145142,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 2810
+    },
+    {
+      "epoch": 3.0870279146141213,
+      "grad_norm": 0.7381885051727295,
+      "learning_rate": 0.0002,
+      "loss": 0.627,
+      "step": 2820
+    },
+    {
+      "epoch": 3.097974822112753,
+      "grad_norm": 0.9197564721107483,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 2830
+    },
+    {
+      "epoch": 3.1089217296113847,
+      "grad_norm": 1.000976800918579,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 2840
+    },
+    {
+      "epoch": 3.1198686371100166,
+      "grad_norm": 0.7559131383895874,
+      "learning_rate": 0.0002,
+      "loss": 0.6631,
+      "step": 2850
+    },
+    {
+      "epoch": 3.130815544608648,
+      "grad_norm": 0.7213780879974365,
+      "learning_rate": 0.0002,
+      "loss": 0.6252,
+      "step": 2860
+    },
+    {
+      "epoch": 3.1417624521072796,
+      "grad_norm": 0.945939838886261,
+      "learning_rate": 0.0002,
+      "loss": 0.6501,
+      "step": 2870
+    },
+    {
+      "epoch": 3.1527093596059115,
+      "grad_norm": 0.7277454137802124,
+      "learning_rate": 0.0002,
+      "loss": 0.6129,
+      "step": 2880
+    },
+    {
+      "epoch": 3.163656267104543,
+      "grad_norm": 0.762026846408844,
+      "learning_rate": 0.0002,
+      "loss": 0.6423,
+      "step": 2890
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 0.6471221446990967,
+      "learning_rate": 0.0002,
+      "loss": 0.5332,
+      "step": 2900
+    },
+    {
+      "epoch": 3.1855500821018063,
+      "grad_norm": 0.6018978357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 2910
+    },
+    {
+      "epoch": 3.196496989600438,
+      "grad_norm": 0.8607320785522461,
+      "learning_rate": 0.0002,
+      "loss": 0.7274,
+      "step": 2920
+    },
+    {
+      "epoch": 3.2074438970990693,
+      "grad_norm": 0.8854126334190369,
+      "learning_rate": 0.0002,
+      "loss": 0.6139,
+      "step": 2930
+    },
+    {
+      "epoch": 3.218390804597701,
+      "grad_norm": 0.6620870232582092,
+      "learning_rate": 0.0002,
+      "loss": 0.6485,
+      "step": 2940
+    },
+    {
+      "epoch": 3.2293377120963327,
+      "grad_norm": 0.7377511858940125,
+      "learning_rate": 0.0002,
+      "loss": 0.6969,
+      "step": 2950
+    },
+    {
+      "epoch": 3.2402846195949646,
+      "grad_norm": 0.7803301811218262,
+      "learning_rate": 0.0002,
+      "loss": 0.6798,
+      "step": 2960
+    },
+    {
+      "epoch": 3.251231527093596,
+      "grad_norm": 0.834061861038208,
+      "learning_rate": 0.0002,
+      "loss": 0.6697,
+      "step": 2970
+    },
+    {
+      "epoch": 3.2621784345922276,
+      "grad_norm": 0.8496041893959045,
+      "learning_rate": 0.0002,
+      "loss": 0.6894,
+      "step": 2980
+    },
+    {
+      "epoch": 3.2731253420908595,
+      "grad_norm": 0.7967984676361084,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 2990
+    },
+    {
+      "epoch": 3.284072249589491,
+      "grad_norm": 1.0207016468048096,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 3000
+    },
+    {
+      "epoch": 3.2950191570881224,
+      "grad_norm": 0.7855771780014038,
+      "learning_rate": 0.0002,
+      "loss": 0.6586,
+      "step": 3010
+    },
+    {
+      "epoch": 3.3059660645867543,
+      "grad_norm": 0.9427546858787537,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 3020
+    },
+    {
+      "epoch": 3.316912972085386,
+      "grad_norm": 0.823542594909668,
+      "learning_rate": 0.0002,
+      "loss": 0.6277,
+      "step": 3030
+    },
+    {
+      "epoch": 3.3278598795840173,
+      "grad_norm": 0.9826635122299194,
+      "learning_rate": 0.0002,
+      "loss": 0.7109,
+      "step": 3040
+    },
+    {
+      "epoch": 3.338806787082649,
+      "grad_norm": 0.7259827852249146,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 3050
+    },
+    {
+      "epoch": 3.3497536945812807,
+      "grad_norm": 0.7774739861488342,
+      "learning_rate": 0.0002,
+      "loss": 0.653,
+      "step": 3060
+    },
+    {
+      "epoch": 3.3607006020799126,
+      "grad_norm": 0.7394293546676636,
+      "learning_rate": 0.0002,
+      "loss": 0.7529,
+      "step": 3070
+    },
+    {
+      "epoch": 3.371647509578544,
+      "grad_norm": 0.9017578959465027,
+      "learning_rate": 0.0002,
+      "loss": 0.5987,
+      "step": 3080
+    },
+    {
+      "epoch": 3.3825944170771756,
+      "grad_norm": 0.7451054453849792,
+      "learning_rate": 0.0002,
+      "loss": 0.6953,
+      "step": 3090
+    },
+    {
+      "epoch": 3.3935413245758075,
+      "grad_norm": 0.7321506142616272,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 3100
+    },
+    {
+      "epoch": 3.404488232074439,
+      "grad_norm": 0.6721828579902649,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3110
+    },
+    {
+      "epoch": 3.4154351395730704,
+      "grad_norm": 0.774022102355957,
+      "learning_rate": 0.0002,
+      "loss": 0.6559,
+      "step": 3120
+    },
+    {
+      "epoch": 3.4263820470717024,
+      "grad_norm": 0.9143537282943726,
+      "learning_rate": 0.0002,
+      "loss": 0.7449,
+      "step": 3130
+    },
+    {
+      "epoch": 3.437328954570334,
+      "grad_norm": 1.226087212562561,
+      "learning_rate": 0.0002,
+      "loss": 0.6899,
+      "step": 3140
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7545496225357056,
+      "learning_rate": 0.0002,
+      "loss": 0.6719,
+      "step": 3150
+    },
+    {
+      "epoch": 3.4592227695675972,
+      "grad_norm": 0.6515635848045349,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 3160
+    },
+    {
+      "epoch": 3.4701696770662287,
+      "grad_norm": 0.9297090172767639,
+      "learning_rate": 0.0002,
+      "loss": 0.6926,
+      "step": 3170
+    },
+    {
+      "epoch": 3.4811165845648606,
+      "grad_norm": 1.0130730867385864,
+      "learning_rate": 0.0002,
+      "loss": 0.6071,
+      "step": 3180
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 0.7654589414596558,
+      "learning_rate": 0.0002,
+      "loss": 0.5959,
+      "step": 3190
+    },
+    {
+      "epoch": 3.5030103995621236,
+      "grad_norm": 0.9954977631568909,
+      "learning_rate": 0.0002,
+      "loss": 0.7401,
+      "step": 3200
+    },
+    {
+      "epoch": 3.5139573070607555,
+      "grad_norm": 0.6027487516403198,
+      "learning_rate": 0.0002,
+      "loss": 0.6661,
+      "step": 3210
+    },
+    {
+      "epoch": 3.524904214559387,
+      "grad_norm": 0.741770327091217,
+      "learning_rate": 0.0002,
+      "loss": 0.6963,
+      "step": 3220
+    },
+    {
+      "epoch": 3.535851122058019,
+      "grad_norm": 1.0534909963607788,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 3230
+    },
+    {
+      "epoch": 3.5467980295566504,
+      "grad_norm": 0.937772274017334,
+      "learning_rate": 0.0002,
+      "loss": 0.6813,
+      "step": 3240
+    },
+    {
+      "epoch": 3.557744937055282,
+      "grad_norm": 0.8504213690757751,
+      "learning_rate": 0.0002,
+      "loss": 0.6681,
+      "step": 3250
+    },
+    {
+      "epoch": 3.5686918445539133,
+      "grad_norm": 0.7755007147789001,
+      "learning_rate": 0.0002,
+      "loss": 0.6436,
+      "step": 3260
+    },
+    {
+      "epoch": 3.5796387520525452,
+      "grad_norm": 1.0193358659744263,
+      "learning_rate": 0.0002,
+      "loss": 0.6213,
+      "step": 3270
+    },
+    {
+      "epoch": 3.5905856595511767,
+      "grad_norm": 0.8440536856651306,
+      "learning_rate": 0.0002,
+      "loss": 0.671,
+      "step": 3280
+    },
+    {
+      "epoch": 3.6015325670498086,
+      "grad_norm": 0.6195939183235168,
+      "learning_rate": 0.0002,
+      "loss": 0.6859,
+      "step": 3290
+    },
+    {
+      "epoch": 3.61247947454844,
+      "grad_norm": 0.8608590960502625,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3300
+    },
+    {
+      "epoch": 3.6234263820470716,
+      "grad_norm": 0.6772327423095703,
+      "learning_rate": 0.0002,
+      "loss": 0.7301,
+      "step": 3310
+    },
+    {
+      "epoch": 3.6343732895457035,
+      "grad_norm": 0.8031839728355408,
+      "learning_rate": 0.0002,
+      "loss": 0.6298,
+      "step": 3320
+    },
+    {
+      "epoch": 3.645320197044335,
+      "grad_norm": 0.6080502271652222,
+      "learning_rate": 0.0002,
+      "loss": 0.7041,
+      "step": 3330
+    },
+    {
+      "epoch": 3.656267104542967,
+      "grad_norm": 0.8007240891456604,
+      "learning_rate": 0.0002,
+      "loss": 0.7431,
+      "step": 3340
+    },
+    {
+      "epoch": 3.6672140120415984,
+      "grad_norm": 0.8060704469680786,
+      "learning_rate": 0.0002,
+      "loss": 0.7446,
+      "step": 3350
+    },
+    {
+      "epoch": 3.67816091954023,
+      "grad_norm": 0.7547586560249329,
+      "learning_rate": 0.0002,
+      "loss": 0.6304,
+      "step": 3360
+    },
+    {
+      "epoch": 3.6891078270388613,
+      "grad_norm": 0.686851978302002,
+      "learning_rate": 0.0002,
+      "loss": 0.7066,
+      "step": 3370
+    },
+    {
+      "epoch": 3.7000547345374932,
+      "grad_norm": 0.9429075717926025,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 3380
+    },
+    {
+      "epoch": 3.7110016420361247,
+      "grad_norm": 0.7283591032028198,
+      "learning_rate": 0.0002,
+      "loss": 0.6673,
+      "step": 3390
+    },
+    {
+      "epoch": 3.7219485495347566,
+      "grad_norm": 0.8323085904121399,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 3400
+    },
+    {
+      "epoch": 3.732895457033388,
+      "grad_norm": 0.8529590964317322,
+      "learning_rate": 0.0002,
+      "loss": 0.7779,
+      "step": 3410
+    },
+    {
+      "epoch": 3.7438423645320196,
+      "grad_norm": 0.731752872467041,
+      "learning_rate": 0.0002,
+      "loss": 0.6555,
+      "step": 3420
+    },
+    {
+      "epoch": 3.7547892720306515,
+      "grad_norm": 0.8572278618812561,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3430
+    },
+    {
+      "epoch": 3.765736179529283,
+      "grad_norm": 0.7408691048622131,
+      "learning_rate": 0.0002,
+      "loss": 0.6215,
+      "step": 3440
+    },
+    {
+      "epoch": 3.776683087027915,
+      "grad_norm": 0.7470445036888123,
+      "learning_rate": 0.0002,
+      "loss": 0.622,
+      "step": 3450
+    },
+    {
+      "epoch": 3.7876299945265464,
+      "grad_norm": 0.6806244254112244,
+      "learning_rate": 0.0002,
+      "loss": 0.7241,
+      "step": 3460
+    },
+    {
+      "epoch": 3.798576902025178,
+      "grad_norm": 0.9129069447517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7739,
+      "step": 3470
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.8717501759529114,
+      "learning_rate": 0.0002,
+      "loss": 0.6826,
+      "step": 3480
+    },
+    {
+      "epoch": 3.8204707170224412,
+      "grad_norm": 0.6761979460716248,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 3490
+    },
+    {
+      "epoch": 3.8314176245210727,
+      "grad_norm": 1.0054380893707275,
+      "learning_rate": 0.0002,
+      "loss": 0.7601,
+      "step": 3500
+    },
+    {
+      "epoch": 3.8423645320197046,
+      "grad_norm": 1.1224009990692139,
+      "learning_rate": 0.0002,
+      "loss": 0.623,
+      "step": 3510
+    },
+    {
+      "epoch": 3.853311439518336,
+      "grad_norm": 0.8997692465782166,
+      "learning_rate": 0.0002,
+      "loss": 0.6918,
+      "step": 3520
+    },
+    {
+      "epoch": 3.8642583470169676,
+      "grad_norm": 1.0086902379989624,
+      "learning_rate": 0.0002,
+      "loss": 0.6357,
+      "step": 3530
+    },
+    {
+      "epoch": 3.8752052545155995,
+      "grad_norm": 0.772739589214325,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 3540
+    },
+    {
+      "epoch": 3.886152162014231,
+      "grad_norm": 1.211774230003357,
+      "learning_rate": 0.0002,
+      "loss": 0.7423,
+      "step": 3550
+    },
+    {
+      "epoch": 3.897099069512863,
+      "grad_norm": 0.9572356939315796,
+      "learning_rate": 0.0002,
+      "loss": 0.7321,
+      "step": 3560
+    },
+    {
+      "epoch": 3.9080459770114944,
+      "grad_norm": 0.7887842655181885,
+      "learning_rate": 0.0002,
+      "loss": 0.6836,
+      "step": 3570
+    },
+    {
+      "epoch": 3.918992884510126,
+      "grad_norm": 0.7308389544487,
+      "learning_rate": 0.0002,
+      "loss": 0.7576,
+      "step": 3580
+    },
+    {
+      "epoch": 3.9299397920087573,
+      "grad_norm": 1.0182650089263916,
+      "learning_rate": 0.0002,
+      "loss": 0.6001,
+      "step": 3590
+    },
+    {
+      "epoch": 3.9408866995073892,
+      "grad_norm": 0.8000147342681885,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 3600
+    },
+    {
+      "epoch": 3.9518336070060207,
+      "grad_norm": 0.7385728359222412,
+      "learning_rate": 0.0002,
+      "loss": 0.6244,
+      "step": 3610
+    },
+    {
+      "epoch": 3.9627805145046526,
+      "grad_norm": 0.9233261942863464,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 3620
+    },
+    {
+      "epoch": 3.973727422003284,
+      "grad_norm": 0.8486751914024353,
+      "learning_rate": 0.0002,
+      "loss": 0.6508,
+      "step": 3630
+    },
+    {
+      "epoch": 3.9846743295019156,
+      "grad_norm": 0.7593663334846497,
+      "learning_rate": 0.0002,
+      "loss": 0.6928,
+      "step": 3640
+    },
+    {
+      "epoch": 3.9956212370005475,
+      "grad_norm": 0.7885415554046631,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 3650
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.250312328338623,
+      "eval_runtime": 46.0842,
+      "eval_samples_per_second": 9.461,
+      "eval_steps_per_second": 1.193,
+      "step": 3654
+    },
+    {
+      "epoch": 4.006568144499179,
+      "grad_norm": 0.6591703295707703,
+      "learning_rate": 0.0002,
+      "loss": 0.5547,
+      "step": 3660
+    },
+    {
+      "epoch": 4.017515051997811,
+      "grad_norm": 1.36927330493927,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 3670
+    },
+    {
+      "epoch": 4.028461959496442,
+      "grad_norm": 0.8106328845024109,
+      "learning_rate": 0.0002,
+      "loss": 0.4466,
+      "step": 3680
+    },
+    {
+      "epoch": 4.039408866995074,
+      "grad_norm": 0.7592712044715881,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 3690
+    },
+    {
+      "epoch": 4.050355774493705,
+      "grad_norm": 0.9518909454345703,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3700
+    },
+    {
+      "epoch": 4.061302681992337,
+      "grad_norm": 0.7805967330932617,
+      "learning_rate": 0.0002,
+      "loss": 0.4638,
+      "step": 3710
+    },
+    {
+      "epoch": 4.072249589490969,
+      "grad_norm": 1.3146334886550903,
+      "learning_rate": 0.0002,
+      "loss": 0.4556,
+      "step": 3720
+    },
+    {
+      "epoch": 4.083196496989601,
+      "grad_norm": 1.1611138582229614,
+      "learning_rate": 0.0002,
+      "loss": 0.5635,
+      "step": 3730
+    },
+    {
+      "epoch": 4.094143404488232,
+      "grad_norm": 0.8173232078552246,
+      "learning_rate": 0.0002,
+      "loss": 0.3845,
+      "step": 3740
+    },
+    {
+      "epoch": 4.105090311986864,
+      "grad_norm": 0.7848323583602905,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 3750
+    },
+    {
+      "epoch": 4.116037219485495,
+      "grad_norm": 1.3183201551437378,
+      "learning_rate": 0.0002,
+      "loss": 0.4519,
+      "step": 3760
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 1.1936529874801636,
+      "learning_rate": 0.0002,
+      "loss": 0.5083,
+      "step": 3770
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1078993082046509,
+      "learning_rate": 0.0002,
+      "loss": 0.5208,
+      "step": 3780
+    },
+    {
+      "epoch": 4.14887794198139,
+      "grad_norm": 1.107743263244629,
+      "learning_rate": 0.0002,
+      "loss": 0.5928,
+      "step": 3790
+    },
+    {
+      "epoch": 4.159824849480022,
+      "grad_norm": 0.7801875472068787,
+      "learning_rate": 0.0002,
+      "loss": 0.5112,
+      "step": 3800
+    },
+    {
+      "epoch": 4.170771756978653,
+      "grad_norm": 1.1328117847442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4896,
+      "step": 3810
+    },
+    {
+      "epoch": 4.181718664477285,
+      "grad_norm": 1.4232193231582642,
+      "learning_rate": 0.0002,
+      "loss": 0.5645,
+      "step": 3820
+    },
+    {
+      "epoch": 4.192665571975917,
+      "grad_norm": 1.557416558265686,
+      "learning_rate": 0.0002,
+      "loss": 0.5049,
+      "step": 3830
+    },
+    {
+      "epoch": 4.203612479474549,
+      "grad_norm": 1.042923092842102,
+      "learning_rate": 0.0002,
+      "loss": 0.4863,
+      "step": 3840
+    },
+    {
+      "epoch": 4.21455938697318,
+      "grad_norm": 1.1801949739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 3850
+    },
+    {
+      "epoch": 4.225506294471812,
+      "grad_norm": 0.9273753762245178,
+      "learning_rate": 0.0002,
+      "loss": 0.5063,
+      "step": 3860
+    },
+    {
+      "epoch": 4.236453201970443,
+      "grad_norm": 0.7681763768196106,
+      "learning_rate": 0.0002,
+      "loss": 0.5542,
+      "step": 3870
+    },
+    {
+      "epoch": 4.2474001094690745,
+      "grad_norm": 0.9840841293334961,
+      "learning_rate": 0.0002,
+      "loss": 0.5971,
+      "step": 3880
+    },
+    {
+      "epoch": 4.258347016967707,
+      "grad_norm": 1.0290725231170654,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 3890
+    },
+    {
+      "epoch": 4.269293924466338,
+      "grad_norm": 0.8059597611427307,
+      "learning_rate": 0.0002,
+      "loss": 0.4288,
+      "step": 3900
+    },
+    {
+      "epoch": 4.28024083196497,
+      "grad_norm": 0.9847467541694641,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 3910
+    },
+    {
+      "epoch": 4.291187739463601,
+      "grad_norm": 1.344044804573059,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 3920
+    },
+    {
+      "epoch": 4.302134646962233,
+      "grad_norm": 0.9174224138259888,
+      "learning_rate": 0.0002,
+      "loss": 0.4966,
+      "step": 3930
+    },
+    {
+      "epoch": 4.313081554460865,
+      "grad_norm": 1.1199711561203003,
+      "learning_rate": 0.0002,
+      "loss": 0.4944,
+      "step": 3940
+    },
+    {
+      "epoch": 4.324028461959497,
+      "grad_norm": 1.0120296478271484,
+      "learning_rate": 0.0002,
+      "loss": 0.4641,
+      "step": 3950
+    },
+    {
+      "epoch": 4.334975369458128,
+      "grad_norm": 1.091811180114746,
+      "learning_rate": 0.0002,
+      "loss": 0.4723,
+      "step": 3960
+    },
+    {
+      "epoch": 4.34592227695676,
+      "grad_norm": 1.0332133769989014,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 3970
+    },
+    {
+      "epoch": 4.356869184455391,
+      "grad_norm": 1.0785295963287354,
+      "learning_rate": 0.0002,
+      "loss": 0.4646,
+      "step": 3980
+    },
+    {
+      "epoch": 4.3678160919540225,
+      "grad_norm": 1.0506969690322876,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 3990
+    },
+    {
+      "epoch": 4.378762999452655,
+      "grad_norm": 1.047560691833496,
+      "learning_rate": 0.0002,
+      "loss": 0.4776,
+      "step": 4000
+    },
+    {
+      "epoch": 4.389709906951286,
+      "grad_norm": 0.9348800778388977,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 4010
+    },
+    {
+      "epoch": 4.400656814449918,
+      "grad_norm": 1.1563059091567993,
+      "learning_rate": 0.0002,
+      "loss": 0.5333,
+      "step": 4020
+    },
+    {
+      "epoch": 4.411603721948549,
+      "grad_norm": 1.001470923423767,
+      "learning_rate": 0.0002,
+      "loss": 0.4952,
+      "step": 4030
+    },
+    {
+      "epoch": 4.422550629447181,
+      "grad_norm": 1.309012532234192,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 4040
+    },
+    {
+      "epoch": 4.433497536945813,
+      "grad_norm": 0.7338925004005432,
+      "learning_rate": 0.0002,
+      "loss": 0.5078,
+      "step": 4050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 1.0398834943771362,
+      "learning_rate": 0.0002,
+      "loss": 0.4632,
+      "step": 4060
+    },
+    {
+      "epoch": 4.455391351943076,
+      "grad_norm": 0.9728689193725586,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4070
+    },
+    {
+      "epoch": 4.466338259441708,
+      "grad_norm": 1.247475028038025,
+      "learning_rate": 0.0002,
+      "loss": 0.4778,
+      "step": 4080
+    },
+    {
+      "epoch": 4.477285166940339,
+      "grad_norm": 1.1084578037261963,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 4090
+    },
+    {
+      "epoch": 4.4882320744389705,
+      "grad_norm": 1.1619318723678589,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 4100
+    },
+    {
+      "epoch": 4.499178981937603,
+      "grad_norm": 1.3456498384475708,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4110
+    },
+    {
+      "epoch": 4.510125889436234,
+      "grad_norm": 0.9372991323471069,
+      "learning_rate": 0.0002,
+      "loss": 0.4876,
+      "step": 4120
+    },
+    {
+      "epoch": 4.521072796934866,
+      "grad_norm": 1.0071815252304077,
+      "learning_rate": 0.0002,
+      "loss": 0.5456,
+      "step": 4130
+    },
+    {
+      "epoch": 4.532019704433497,
+      "grad_norm": 1.190344214439392,
+      "learning_rate": 0.0002,
+      "loss": 0.5589,
+      "step": 4140
+    },
+    {
+      "epoch": 4.542966611932129,
+      "grad_norm": 0.9480887055397034,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 4150
+    },
+    {
+      "epoch": 4.553913519430761,
+      "grad_norm": 1.0252189636230469,
+      "learning_rate": 0.0002,
+      "loss": 0.5229,
+      "step": 4160
+    },
+    {
+      "epoch": 4.564860426929393,
+      "grad_norm": 0.7142013311386108,
+      "learning_rate": 0.0002,
+      "loss": 0.5253,
+      "step": 4170
+    },
+    {
+      "epoch": 4.575807334428024,
+      "grad_norm": 0.8937426805496216,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 4180
+    },
+    {
+      "epoch": 4.586754241926656,
+      "grad_norm": 0.8885005116462708,
+      "learning_rate": 0.0002,
+      "loss": 0.4773,
+      "step": 4190
+    },
+    {
+      "epoch": 4.597701149425287,
+      "grad_norm": 1.337663173675537,
+      "learning_rate": 0.0002,
+      "loss": 0.4858,
+      "step": 4200
+    },
+    {
+      "epoch": 4.6086480569239185,
+      "grad_norm": 1.0475375652313232,
+      "learning_rate": 0.0002,
+      "loss": 0.5247,
+      "step": 4210
+    },
+    {
+      "epoch": 4.619594964422551,
+      "grad_norm": 1.0081088542938232,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 4220
+    },
+    {
+      "epoch": 4.630541871921182,
+      "grad_norm": 0.7527595162391663,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 4230
+    },
+    {
+      "epoch": 4.641488779419814,
+      "grad_norm": 1.55559241771698,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 4240
+    },
+    {
+      "epoch": 4.652435686918445,
+      "grad_norm": 0.7967379689216614,
+      "learning_rate": 0.0002,
+      "loss": 0.5468,
+      "step": 4250
+    },
+    {
+      "epoch": 4.663382594417077,
+      "grad_norm": 0.898368775844574,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 4260
+    },
+    {
+      "epoch": 4.674329501915709,
+      "grad_norm": 1.1940776109695435,
+      "learning_rate": 0.0002,
+      "loss": 0.4706,
+      "step": 4270
+    },
+    {
+      "epoch": 4.685276409414341,
+      "grad_norm": 1.1817092895507812,
+      "learning_rate": 0.0002,
+      "loss": 0.5121,
+      "step": 4280
+    },
+    {
+      "epoch": 4.696223316912972,
+      "grad_norm": 0.9041520357131958,
+      "learning_rate": 0.0002,
+      "loss": 0.5758,
+      "step": 4290
+    },
+    {
+      "epoch": 4.707170224411604,
+      "grad_norm": 1.1280102729797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5851,
+      "step": 4300
+    },
+    {
+      "epoch": 4.718117131910235,
+      "grad_norm": 1.357689619064331,
+      "learning_rate": 0.0002,
+      "loss": 0.4891,
+      "step": 4310
+    },
+    {
+      "epoch": 4.7290640394088665,
+      "grad_norm": 1.056633472442627,
+      "learning_rate": 0.0002,
+      "loss": 0.4704,
+      "step": 4320
+    },
+    {
+      "epoch": 4.740010946907499,
+      "grad_norm": 1.6520427465438843,
+      "learning_rate": 0.0002,
+      "loss": 0.5488,
+      "step": 4330
+    },
+    {
+      "epoch": 4.75095785440613,
+      "grad_norm": 1.153200626373291,
+      "learning_rate": 0.0002,
+      "loss": 0.5131,
+      "step": 4340
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.9346241354942322,
+      "learning_rate": 0.0002,
+      "loss": 0.539,
+      "step": 4350
+    },
+    {
+      "epoch": 4.772851669403393,
+      "grad_norm": 0.8628455996513367,
+      "learning_rate": 0.0002,
+      "loss": 0.4941,
+      "step": 4360
+    },
+    {
+      "epoch": 4.783798576902025,
+      "grad_norm": 1.3843916654586792,
+      "learning_rate": 0.0002,
+      "loss": 0.5167,
+      "step": 4370
+    },
+    {
+      "epoch": 4.794745484400657,
+      "grad_norm": 1.035574197769165,
+      "learning_rate": 0.0002,
+      "loss": 0.4683,
+      "step": 4380
+    },
+    {
+      "epoch": 4.805692391899289,
+      "grad_norm": 1.1868361234664917,
+      "learning_rate": 0.0002,
+      "loss": 0.5162,
+      "step": 4390
+    },
+    {
+      "epoch": 4.81663929939792,
+      "grad_norm": 1.1307647228240967,
+      "learning_rate": 0.0002,
+      "loss": 0.534,
+      "step": 4400
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 0.9787724614143372,
+      "learning_rate": 0.0002,
+      "loss": 0.5567,
+      "step": 4410
+    },
+    {
+      "epoch": 4.838533114395183,
+      "grad_norm": 1.0473824739456177,
+      "learning_rate": 0.0002,
+      "loss": 0.5185,
+      "step": 4420
+    },
+    {
+      "epoch": 4.8494800218938146,
+      "grad_norm": 1.069069504737854,
+      "learning_rate": 0.0002,
+      "loss": 0.6285,
+      "step": 4430
+    },
+    {
+      "epoch": 4.860426929392447,
+      "grad_norm": 1.4305680990219116,
+      "learning_rate": 0.0002,
+      "loss": 0.5267,
+      "step": 4440
+    },
+    {
+      "epoch": 4.871373836891078,
+      "grad_norm": 1.3679203987121582,
+      "learning_rate": 0.0002,
+      "loss": 0.5947,
+      "step": 4450
+    },
+    {
+      "epoch": 4.88232074438971,
+      "grad_norm": 0.8997844457626343,
+      "learning_rate": 0.0002,
+      "loss": 0.5135,
+      "step": 4460
+    },
+    {
+      "epoch": 4.893267651888341,
+      "grad_norm": 1.2758110761642456,
+      "learning_rate": 0.0002,
+      "loss": 0.5312,
+      "step": 4470
+    },
+    {
+      "epoch": 4.904214559386973,
+      "grad_norm": 0.8819465637207031,
+      "learning_rate": 0.0002,
+      "loss": 0.4914,
+      "step": 4480
+    },
+    {
+      "epoch": 4.915161466885605,
+      "grad_norm": 1.08329439163208,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 4490
+    },
+    {
+      "epoch": 4.926108374384237,
+      "grad_norm": 1.083461046218872,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 4500
+    },
+    {
+      "epoch": 4.937055281882868,
+      "grad_norm": 1.2387723922729492,
+      "learning_rate": 0.0002,
+      "loss": 0.5433,
+      "step": 4510
+    },
+    {
+      "epoch": 4.9480021893815,
+      "grad_norm": 0.8262293934822083,
+      "learning_rate": 0.0002,
+      "loss": 0.5624,
+      "step": 4520
+    },
+    {
+      "epoch": 4.958949096880131,
+      "grad_norm": 1.2325191497802734,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 4530
+    },
+    {
+      "epoch": 4.9698960043787626,
+      "grad_norm": 1.024614930152893,
+      "learning_rate": 0.0002,
+      "loss": 0.5452,
+      "step": 4540
+    },
+    {
+      "epoch": 4.980842911877395,
+      "grad_norm": 1.3007521629333496,
+      "learning_rate": 0.0002,
+      "loss": 0.4752,
+      "step": 4550
+    },
+    {
+      "epoch": 4.991789819376026,
+      "grad_norm": 0.9823828339576721,
+      "learning_rate": 0.0002,
+      "loss": 0.4943,
+      "step": 4560
+    },
+    {
+      "epoch": 4.999452654625069,
+      "eval_loss": 1.3920727968215942,
+      "eval_runtime": 46.0764,
+      "eval_samples_per_second": 9.463,
+      "eval_steps_per_second": 1.194,
+      "step": 4567
+    },
+    {
+      "epoch": 5.002736726874658,
+      "grad_norm": 1.1478906869888306,
+      "learning_rate": 0.0002,
+      "loss": 0.545,
+      "step": 4570
+    },
+    {
+      "epoch": 5.013683634373289,
+      "grad_norm": 1.0533705949783325,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 4580
+    },
+    {
+      "epoch": 5.024630541871921,
+      "grad_norm": 1.268900752067566,
+      "learning_rate": 0.0002,
+      "loss": 0.3313,
+      "step": 4590
+    },
+    {
+      "epoch": 5.035577449370553,
+      "grad_norm": 1.222652554512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3482,
+      "step": 4600
+    },
+    {
+      "epoch": 5.046524356869185,
+      "grad_norm": 1.5093127489089966,
+      "learning_rate": 0.0002,
+      "loss": 0.3195,
+      "step": 4610
+    },
+    {
+      "epoch": 5.057471264367816,
+      "grad_norm": 1.2372499704360962,
+      "learning_rate": 0.0002,
+      "loss": 0.3569,
+      "step": 4620
+    },
+    {
+      "epoch": 5.068418171866448,
+      "grad_norm": 0.8422666192054749,
+      "learning_rate": 0.0002,
+      "loss": 0.3206,
+      "step": 4630
+    },
+    {
+      "epoch": 5.079365079365079,
+      "grad_norm": 1.1451770067214966,
+      "learning_rate": 0.0002,
+      "loss": 0.3115,
+      "step": 4640
+    },
+    {
+      "epoch": 5.090311986863711,
+      "grad_norm": 1.2074557542800903,
+      "learning_rate": 0.0002,
+      "loss": 0.3305,
+      "step": 4650
+    },
+    {
+      "epoch": 5.101258894362343,
+      "grad_norm": 1.429150104522705,
+      "learning_rate": 0.0002,
+      "loss": 0.3012,
+      "step": 4660
+    },
+    {
+      "epoch": 5.112205801860974,
+      "grad_norm": 1.0353610515594482,
+      "learning_rate": 0.0002,
+      "loss": 0.3229,
+      "step": 4670
+    },
+    {
+      "epoch": 5.123152709359606,
+      "grad_norm": 1.2845979928970337,
+      "learning_rate": 0.0002,
+      "loss": 0.402,
+      "step": 4680
+    },
+    {
+      "epoch": 5.134099616858237,
+      "grad_norm": 1.3790186643600464,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 4690
+    },
+    {
+      "epoch": 5.145046524356869,
+      "grad_norm": 1.3182239532470703,
+      "learning_rate": 0.0002,
+      "loss": 0.2951,
+      "step": 4700
+    },
+    {
+      "epoch": 5.155993431855501,
+      "grad_norm": 1.5249626636505127,
+      "learning_rate": 0.0002,
+      "loss": 0.4074,
+      "step": 4710
+    },
+    {
+      "epoch": 5.166940339354133,
+      "grad_norm": 1.2492733001708984,
+      "learning_rate": 0.0002,
+      "loss": 0.3703,
+      "step": 4720
+    },
+    {
+      "epoch": 5.177887246852764,
+      "grad_norm": 1.4455480575561523,
+      "learning_rate": 0.0002,
+      "loss": 0.3411,
+      "step": 4730
+    },
+    {
+      "epoch": 5.188834154351396,
+      "grad_norm": 1.2191482782363892,
+      "learning_rate": 0.0002,
+      "loss": 0.3996,
+      "step": 4740
+    },
+    {
+      "epoch": 5.199781061850027,
+      "grad_norm": 1.4707951545715332,
+      "learning_rate": 0.0002,
+      "loss": 0.3785,
+      "step": 4750
+    },
+    {
+      "epoch": 5.210727969348659,
+      "grad_norm": 1.3473678827285767,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4760
+    },
+    {
+      "epoch": 5.221674876847291,
+      "grad_norm": 1.0479670763015747,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4770
+    },
+    {
+      "epoch": 5.232621784345922,
+      "grad_norm": 1.299096703529358,
+      "learning_rate": 0.0002,
+      "loss": 0.3976,
+      "step": 4780
+    },
+    {
+      "epoch": 5.243568691844554,
+      "grad_norm": 1.2820168733596802,
+      "learning_rate": 0.0002,
+      "loss": 0.3266,
+      "step": 4790
+    },
+    {
+      "epoch": 5.254515599343185,
+      "grad_norm": 1.3818004131317139,
+      "learning_rate": 0.0002,
+      "loss": 0.3347,
+      "step": 4800
+    },
+    {
+      "epoch": 5.265462506841817,
+      "grad_norm": 1.2898736000061035,
+      "learning_rate": 0.0002,
+      "loss": 0.3761,
+      "step": 4810
+    },
+    {
+      "epoch": 5.276409414340449,
+      "grad_norm": 1.1761468648910522,
+      "learning_rate": 0.0002,
+      "loss": 0.3694,
+      "step": 4820
+    },
+    {
+      "epoch": 5.287356321839081,
+      "grad_norm": 1.7155952453613281,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4830
+    },
+    {
+      "epoch": 5.298303229337712,
+      "grad_norm": 0.9103642106056213,
+      "learning_rate": 0.0002,
+      "loss": 0.322,
+      "step": 4840
+    },
+    {
+      "epoch": 5.309250136836344,
+      "grad_norm": 1.013015627861023,
+      "learning_rate": 0.0002,
+      "loss": 0.3516,
+      "step": 4850
+    },
+    {
+      "epoch": 5.320197044334975,
+      "grad_norm": 1.390471339225769,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 4860
+    },
+    {
+      "epoch": 5.331143951833607,
+      "grad_norm": 1.129770278930664,
+      "learning_rate": 0.0002,
+      "loss": 0.4098,
+      "step": 4870
+    },
+    {
+      "epoch": 5.342090859332239,
+      "grad_norm": 1.1461067199707031,
+      "learning_rate": 0.0002,
+      "loss": 0.4227,
+      "step": 4880
+    },
+    {
+      "epoch": 5.35303776683087,
+      "grad_norm": 1.3587424755096436,
+      "learning_rate": 0.0002,
+      "loss": 0.288,
+      "step": 4890
+    },
+    {
+      "epoch": 5.363984674329502,
+      "grad_norm": 1.6897879838943481,
+      "learning_rate": 0.0002,
+      "loss": 0.3604,
+      "step": 4900
+    },
+    {
+      "epoch": 5.374931581828133,
+      "grad_norm": 0.9298055768013,
+      "learning_rate": 0.0002,
+      "loss": 0.3887,
+      "step": 4910
+    },
+    {
+      "epoch": 5.385878489326765,
+      "grad_norm": 1.0006917715072632,
+      "learning_rate": 0.0002,
+      "loss": 0.3371,
+      "step": 4920
+    },
+    {
+      "epoch": 5.396825396825397,
+      "grad_norm": 1.232581377029419,
+      "learning_rate": 0.0002,
+      "loss": 0.3992,
+      "step": 4930
+    },
+    {
+      "epoch": 5.407772304324029,
+      "grad_norm": 1.0822620391845703,
+      "learning_rate": 0.0002,
+      "loss": 0.3456,
+      "step": 4940
+    },
+    {
+      "epoch": 5.41871921182266,
+      "grad_norm": 1.3648720979690552,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 4950
+    },
+    {
+      "epoch": 5.429666119321292,
+      "grad_norm": 1.3220354318618774,
+      "learning_rate": 0.0002,
+      "loss": 0.3959,
+      "step": 4960
+    },
+    {
+      "epoch": 5.440613026819923,
+      "grad_norm": 1.1106271743774414,
+      "learning_rate": 0.0002,
+      "loss": 0.3278,
+      "step": 4970
+    },
+    {
+      "epoch": 5.451559934318555,
+      "grad_norm": 1.6058908700942993,
+      "learning_rate": 0.0002,
+      "loss": 0.3812,
+      "step": 4980
+    },
+    {
+      "epoch": 5.462506841817187,
+      "grad_norm": 1.1065930128097534,
+      "learning_rate": 0.0002,
+      "loss": 0.3905,
+      "step": 4990
+    },
+    {
+      "epoch": 5.473453749315818,
+      "grad_norm": 1.3896466493606567,
+      "learning_rate": 0.0002,
+      "loss": 0.4058,
+      "step": 5000
+    },
+    {
+      "epoch": 5.48440065681445,
+      "grad_norm": 1.0437148809432983,
+      "learning_rate": 0.0002,
+      "loss": 0.4122,
+      "step": 5010
+    },
+    {
+      "epoch": 5.495347564313081,
+      "grad_norm": 1.2347718477249146,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 5020
+    },
+    {
+      "epoch": 5.506294471811713,
+      "grad_norm": 1.1174284219741821,
+      "learning_rate": 0.0002,
+      "loss": 0.3586,
+      "step": 5030
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2580941915512085,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 5040
+    },
+    {
+      "epoch": 5.528188286808977,
+      "grad_norm": 1.451090931892395,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 5050
+    },
+    {
+      "epoch": 5.539135194307608,
+      "grad_norm": 1.4688365459442139,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 5060
+    },
+    {
+      "epoch": 5.55008210180624,
+      "grad_norm": 1.1625734567642212,
+      "learning_rate": 0.0002,
+      "loss": 0.4431,
+      "step": 5070
+    },
+    {
+      "epoch": 5.561029009304871,
+      "grad_norm": 0.9332265257835388,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 5080
+    },
+    {
+      "epoch": 5.571975916803503,
+      "grad_norm": 1.5635273456573486,
+      "learning_rate": 0.0002,
+      "loss": 0.4,
+      "step": 5090
+    },
+    {
+      "epoch": 5.582922824302135,
+      "grad_norm": 1.3420509099960327,
+      "learning_rate": 0.0002,
+      "loss": 0.3651,
+      "step": 5100
+    },
+    {
+      "epoch": 5.593869731800766,
+      "grad_norm": 1.5826557874679565,
+      "learning_rate": 0.0002,
+      "loss": 0.3717,
+      "step": 5110
+    },
+    {
+      "epoch": 5.604816639299398,
+      "grad_norm": 1.5737065076828003,
+      "learning_rate": 0.0002,
+      "loss": 0.4256,
+      "step": 5120
+    },
+    {
+      "epoch": 5.615763546798029,
+      "grad_norm": 1.3812499046325684,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 5130
+    },
+    {
+      "epoch": 5.626710454296661,
+      "grad_norm": 1.362833023071289,
+      "learning_rate": 0.0002,
+      "loss": 0.3891,
+      "step": 5140
+    },
+    {
+      "epoch": 5.637657361795293,
+      "grad_norm": 1.7667874097824097,
+      "learning_rate": 0.0002,
+      "loss": 0.455,
+      "step": 5150
+    },
+    {
+      "epoch": 5.648604269293925,
+      "grad_norm": 1.2661789655685425,
+      "learning_rate": 0.0002,
+      "loss": 0.4264,
+      "step": 5160
+    },
+    {
+      "epoch": 5.659551176792556,
+      "grad_norm": 1.2076870203018188,
+      "learning_rate": 0.0002,
+      "loss": 0.3261,
+      "step": 5170
+    },
+    {
+      "epoch": 5.670498084291188,
+      "grad_norm": 1.2431524991989136,
+      "learning_rate": 0.0002,
+      "loss": 0.372,
+      "step": 5180
+    },
+    {
+      "epoch": 5.681444991789819,
+      "grad_norm": 1.2216639518737793,
+      "learning_rate": 0.0002,
+      "loss": 0.4092,
+      "step": 5190
+    },
+    {
+      "epoch": 5.692391899288451,
+      "grad_norm": 0.9259352684020996,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5200
+    },
+    {
+      "epoch": 5.703338806787083,
+      "grad_norm": 1.7929338216781616,
+      "learning_rate": 0.0002,
+      "loss": 0.3875,
+      "step": 5210
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 1.4048460721969604,
+      "learning_rate": 0.0002,
+      "loss": 0.4424,
+      "step": 5220
+    },
+    {
+      "epoch": 5.725232621784346,
+      "grad_norm": 1.306874394416809,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 5230
+    },
+    {
+      "epoch": 5.736179529282977,
+      "grad_norm": 1.3137940168380737,
+      "learning_rate": 0.0002,
+      "loss": 0.3889,
+      "step": 5240
+    },
+    {
+      "epoch": 5.747126436781609,
+      "grad_norm": 1.1376476287841797,
+      "learning_rate": 0.0002,
+      "loss": 0.4804,
+      "step": 5250
+    },
+    {
+      "epoch": 5.758073344280241,
+      "grad_norm": 1.450939416885376,
+      "learning_rate": 0.0002,
+      "loss": 0.377,
+      "step": 5260
+    },
+    {
+      "epoch": 5.769020251778873,
+      "grad_norm": 0.983195960521698,
+      "learning_rate": 0.0002,
+      "loss": 0.4732,
+      "step": 5270
+    },
+    {
+      "epoch": 5.779967159277504,
+      "grad_norm": 1.66558837890625,
+      "learning_rate": 0.0002,
+      "loss": 0.4041,
+      "step": 5280
+    },
+    {
+      "epoch": 5.790914066776136,
+      "grad_norm": 0.9789204597473145,
+      "learning_rate": 0.0002,
+      "loss": 0.3643,
+      "step": 5290
+    },
+    {
+      "epoch": 5.801860974274767,
+      "grad_norm": 1.2110556364059448,
+      "learning_rate": 0.0002,
+      "loss": 0.3776,
+      "step": 5300
+    },
+    {
+      "epoch": 5.812807881773399,
+      "grad_norm": 1.3799304962158203,
+      "learning_rate": 0.0002,
+      "loss": 0.4049,
+      "step": 5310
+    },
+    {
+      "epoch": 5.823754789272031,
+      "grad_norm": 1.0570626258850098,
+      "learning_rate": 0.0002,
+      "loss": 0.4362,
+      "step": 5320
+    },
+    {
+      "epoch": 5.834701696770662,
+      "grad_norm": 1.4654436111450195,
+      "learning_rate": 0.0002,
+      "loss": 0.4716,
+      "step": 5330
+    },
+    {
+      "epoch": 5.845648604269294,
+      "grad_norm": 1.5216940641403198,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 5340
+    },
+    {
+      "epoch": 5.856595511767925,
+      "grad_norm": 1.018646001815796,
+      "learning_rate": 0.0002,
+      "loss": 0.3848,
+      "step": 5350
+    },
+    {
+      "epoch": 5.867542419266557,
+      "grad_norm": 1.028951644897461,
+      "learning_rate": 0.0002,
+      "loss": 0.3705,
+      "step": 5360
+    },
+    {
+      "epoch": 5.878489326765189,
+      "grad_norm": 2.571263313293457,
+      "learning_rate": 0.0002,
+      "loss": 0.4213,
+      "step": 5370
+    },
+    {
+      "epoch": 5.889436234263821,
+      "grad_norm": 1.3323984146118164,
+      "learning_rate": 0.0002,
+      "loss": 0.3647,
+      "step": 5380
+    },
+    {
+      "epoch": 5.900383141762452,
+      "grad_norm": 1.4317777156829834,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 5390
+    },
+    {
+      "epoch": 5.911330049261084,
+      "grad_norm": 1.4289140701293945,
+      "learning_rate": 0.0002,
+      "loss": 0.4254,
+      "step": 5400
+    },
+    {
+      "epoch": 5.922276956759715,
+      "grad_norm": 1.3130780458450317,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 5410
+    },
+    {
+      "epoch": 5.933223864258347,
+      "grad_norm": 1.3979902267456055,
+      "learning_rate": 0.0002,
+      "loss": 0.4025,
+      "step": 5420
+    },
+    {
+      "epoch": 5.944170771756979,
+      "grad_norm": 1.1827352046966553,
+      "learning_rate": 0.0002,
+      "loss": 0.3997,
+      "step": 5430
+    },
+    {
+      "epoch": 5.95511767925561,
+      "grad_norm": 1.1672080755233765,
+      "learning_rate": 0.0002,
+      "loss": 0.4163,
+      "step": 5440
+    },
+    {
+      "epoch": 5.966064586754242,
+      "grad_norm": 1.0949620008468628,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 5450
+    },
+    {
+      "epoch": 5.977011494252873,
+      "grad_norm": 1.3183925151824951,
+      "learning_rate": 0.0002,
+      "loss": 0.4219,
+      "step": 5460
+    },
+    {
+      "epoch": 5.987958401751505,
+      "grad_norm": 1.096198320388794,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 5470
+    },
+    {
+      "epoch": 5.998905309250137,
+      "grad_norm": 1.2601423263549805,
+      "learning_rate": 0.0002,
+      "loss": 0.3886,
+      "step": 5480
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.611358880996704,
+      "eval_runtime": 46.0638,
+      "eval_samples_per_second": 9.465,
+      "eval_steps_per_second": 1.194,
+      "step": 5481
+    },
+    {
+      "epoch": 6.009852216748769,
+      "grad_norm": 0.9854364991188049,
+      "learning_rate": 0.0002,
+      "loss": 0.2616,
+      "step": 5490
+    },
+    {
+      "epoch": 6.0207991242474,
+      "grad_norm": 1.8073689937591553,
+      "learning_rate": 0.0002,
+      "loss": 0.2412,
+      "step": 5500
+    },
+    {
+      "epoch": 6.031746031746032,
+      "grad_norm": 1.1852164268493652,
+      "learning_rate": 0.0002,
+      "loss": 0.2317,
+      "step": 5510
+    },
+    {
+      "epoch": 6.042692939244663,
+      "grad_norm": 1.0937914848327637,
+      "learning_rate": 0.0002,
+      "loss": 0.224,
+      "step": 5520
+    },
+    {
+      "epoch": 6.053639846743295,
+      "grad_norm": 0.7411194443702698,
+      "learning_rate": 0.0002,
+      "loss": 0.2473,
+      "step": 5530
+    },
+    {
+      "epoch": 6.064586754241927,
+      "grad_norm": 1.552127480506897,
+      "learning_rate": 0.0002,
+      "loss": 0.2846,
+      "step": 5540
+    },
+    {
+      "epoch": 6.075533661740558,
+      "grad_norm": 1.0465604066848755,
+      "learning_rate": 0.0002,
+      "loss": 0.2639,
+      "step": 5550
+    },
+    {
+      "epoch": 6.08648056923919,
+      "grad_norm": 1.4008121490478516,
+      "learning_rate": 0.0002,
+      "loss": 0.2696,
+      "step": 5560
+    },
+    {
+      "epoch": 6.097427476737821,
+      "grad_norm": 1.7049046754837036,
+      "learning_rate": 0.0002,
+      "loss": 0.3049,
+      "step": 5570
+    },
+    {
+      "epoch": 6.108374384236453,
+      "grad_norm": 1.111151933670044,
+      "learning_rate": 0.0002,
+      "loss": 0.263,
+      "step": 5580
+    },
+    {
+      "epoch": 6.119321291735085,
+      "grad_norm": 1.4271087646484375,
+      "learning_rate": 0.0002,
+      "loss": 0.2816,
+      "step": 5590
+    },
+    {
+      "epoch": 6.130268199233717,
+      "grad_norm": 1.3917373418807983,
+      "learning_rate": 0.0002,
+      "loss": 0.2878,
+      "step": 5600
+    },
+    {
+      "epoch": 6.141215106732348,
+      "grad_norm": 1.013689637184143,
+      "learning_rate": 0.0002,
+      "loss": 0.2482,
+      "step": 5610
+    },
+    {
+      "epoch": 6.15216201423098,
+      "grad_norm": 1.342645525932312,
+      "learning_rate": 0.0002,
+      "loss": 0.2841,
+      "step": 5620
+    },
+    {
+      "epoch": 6.163108921729611,
+      "grad_norm": 1.4480562210083008,
+      "learning_rate": 0.0002,
+      "loss": 0.2335,
+      "step": 5630
+    },
+    {
+      "epoch": 6.174055829228243,
+      "grad_norm": 1.2483175992965698,
+      "learning_rate": 0.0002,
+      "loss": 0.2696,
+      "step": 5640
+    },
+    {
+      "epoch": 6.185002736726875,
+      "grad_norm": 1.2944550514221191,
+      "learning_rate": 0.0002,
+      "loss": 0.2656,
+      "step": 5650
+    },
+    {
+      "epoch": 6.195949644225506,
+      "grad_norm": 1.264142632484436,
+      "learning_rate": 0.0002,
+      "loss": 0.2704,
+      "step": 5660
+    },
+    {
+      "epoch": 6.206896551724138,
+      "grad_norm": 1.2068781852722168,
+      "learning_rate": 0.0002,
+      "loss": 0.2971,
+      "step": 5670
+    },
+    {
+      "epoch": 6.217843459222769,
+      "grad_norm": 1.0401629209518433,
+      "learning_rate": 0.0002,
+      "loss": 0.2882,
+      "step": 5680
+    },
+    {
+      "epoch": 6.228790366721401,
+      "grad_norm": 1.2054402828216553,
+      "learning_rate": 0.0002,
+      "loss": 0.3022,
+      "step": 5690
+    },
+    {
+      "epoch": 6.239737274220033,
+      "grad_norm": 1.1278687715530396,
+      "learning_rate": 0.0002,
+      "loss": 0.2949,
+      "step": 5700
+    },
+    {
+      "epoch": 6.250684181718665,
+      "grad_norm": 1.24592125415802,
+      "learning_rate": 0.0002,
+      "loss": 0.2477,
+      "step": 5710
+    },
+    {
+      "epoch": 6.261631089217296,
+      "grad_norm": 1.2686697244644165,
+      "learning_rate": 0.0002,
+      "loss": 0.246,
+      "step": 5720
+    },
+    {
+      "epoch": 6.272577996715928,
+      "grad_norm": 1.1836518049240112,
+      "learning_rate": 0.0002,
+      "loss": 0.2974,
+      "step": 5730
+    },
+    {
+      "epoch": 6.283524904214559,
+      "grad_norm": 1.387752890586853,
+      "learning_rate": 0.0002,
+      "loss": 0.2963,
+      "step": 5740
+    },
+    {
+      "epoch": 6.294471811713191,
+      "grad_norm": 1.9390363693237305,
+      "learning_rate": 0.0002,
+      "loss": 0.2961,
+      "step": 5750
+    },
+    {
+      "epoch": 6.305418719211823,
+      "grad_norm": 1.2919824123382568,
+      "learning_rate": 0.0002,
+      "loss": 0.2765,
+      "step": 5760
+    },
+    {
+      "epoch": 6.316365626710454,
+      "grad_norm": 1.2793965339660645,
+      "learning_rate": 0.0002,
+      "loss": 0.2898,
+      "step": 5770
+    },
+    {
+      "epoch": 6.327312534209086,
+      "grad_norm": 1.5486980676651,
+      "learning_rate": 0.0002,
+      "loss": 0.2786,
+      "step": 5780
+    },
+    {
+      "epoch": 6.338259441707717,
+      "grad_norm": 1.2757408618927002,
+      "learning_rate": 0.0002,
+      "loss": 0.2684,
+      "step": 5790
+    },
+    {
+      "epoch": 6.349206349206349,
+      "grad_norm": 1.3245713710784912,
+      "learning_rate": 0.0002,
+      "loss": 0.2841,
+      "step": 5800
+    },
+    {
+      "epoch": 6.360153256704981,
+      "grad_norm": 1.6262527704238892,
+      "learning_rate": 0.0002,
+      "loss": 0.3096,
+      "step": 5810
+    },
+    {
+      "epoch": 6.371100164203613,
+      "grad_norm": 1.465224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.3219,
+      "step": 5820
+    },
+    {
+      "epoch": 6.382047071702244,
+      "grad_norm": 1.437408447265625,
+      "learning_rate": 0.0002,
+      "loss": 0.2703,
+      "step": 5830
+    },
+    {
+      "epoch": 6.392993979200876,
+      "grad_norm": 1.3094626665115356,
+      "learning_rate": 0.0002,
+      "loss": 0.3012,
+      "step": 5840
+    },
+    {
+      "epoch": 6.403940886699507,
+      "grad_norm": 1.6717544794082642,
+      "learning_rate": 0.0002,
+      "loss": 0.2991,
+      "step": 5850
+    },
+    {
+      "epoch": 6.414887794198139,
+      "grad_norm": 1.1023344993591309,
+      "learning_rate": 0.0002,
+      "loss": 0.2892,
+      "step": 5860
+    },
+    {
+      "epoch": 6.425834701696771,
+      "grad_norm": 1.2397106885910034,
+      "learning_rate": 0.0002,
+      "loss": 0.3078,
+      "step": 5870
+    },
+    {
+      "epoch": 6.436781609195402,
+      "grad_norm": 1.6139185428619385,
+      "learning_rate": 0.0002,
+      "loss": 0.2984,
+      "step": 5880
+    },
+    {
+      "epoch": 6.447728516694034,
+      "grad_norm": 1.3164576292037964,
+      "learning_rate": 0.0002,
+      "loss": 0.2353,
+      "step": 5890
+    },
+    {
+      "epoch": 6.458675424192665,
+      "grad_norm": 1.3317217826843262,
+      "learning_rate": 0.0002,
+      "loss": 0.2772,
+      "step": 5900
+    },
+    {
+      "epoch": 6.469622331691297,
+      "grad_norm": 1.215008020401001,
+      "learning_rate": 0.0002,
+      "loss": 0.2555,
+      "step": 5910
+    },
+    {
+      "epoch": 6.480569239189929,
+      "grad_norm": 1.625672698020935,
+      "learning_rate": 0.0002,
+      "loss": 0.2715,
+      "step": 5920
+    },
+    {
+      "epoch": 6.491516146688561,
+      "grad_norm": 1.1262489557266235,
+      "learning_rate": 0.0002,
+      "loss": 0.2938,
+      "step": 5930
+    },
+    {
+      "epoch": 6.502463054187192,
+      "grad_norm": 1.447100281715393,
+      "learning_rate": 0.0002,
+      "loss": 0.2921,
+      "step": 5940
+    },
+    {
+      "epoch": 6.513409961685824,
+      "grad_norm": 1.3306448459625244,
+      "learning_rate": 0.0002,
+      "loss": 0.3059,
+      "step": 5950
+    },
+    {
+      "epoch": 6.524356869184455,
+      "grad_norm": 1.307732105255127,
+      "learning_rate": 0.0002,
+      "loss": 0.2922,
+      "step": 5960
+    },
+    {
+      "epoch": 6.535303776683087,
+      "grad_norm": 1.1851097345352173,
+      "learning_rate": 0.0002,
+      "loss": 0.2891,
+      "step": 5970
+    },
+    {
+      "epoch": 6.546250684181719,
+      "grad_norm": 1.462816596031189,
+      "learning_rate": 0.0002,
+      "loss": 0.2859,
+      "step": 5980
+    },
+    {
+      "epoch": 6.55719759168035,
+      "grad_norm": 1.2324728965759277,
+      "learning_rate": 0.0002,
+      "loss": 0.2698,
+      "step": 5990
+    },
+    {
+      "epoch": 6.568144499178982,
+      "grad_norm": 1.3627429008483887,
+      "learning_rate": 0.0002,
+      "loss": 0.2672,
+      "step": 6000
+    },
+    {
+      "epoch": 6.579091406677613,
+      "grad_norm": 1.94977867603302,
+      "learning_rate": 0.0002,
+      "loss": 0.3182,
+      "step": 6010
+    },
+    {
+      "epoch": 6.590038314176245,
+      "grad_norm": 1.459844946861267,
+      "learning_rate": 0.0002,
+      "loss": 0.3183,
+      "step": 6020
+    },
+    {
+      "epoch": 6.600985221674877,
+      "grad_norm": 1.4454325437545776,
+      "learning_rate": 0.0002,
+      "loss": 0.3142,
+      "step": 6030
+    },
+    {
+      "epoch": 6.611932129173509,
+      "grad_norm": 1.4245165586471558,
+      "learning_rate": 0.0002,
+      "loss": 0.269,
+      "step": 6040
+    },
+    {
+      "epoch": 6.62287903667214,
+      "grad_norm": 1.195803165435791,
+      "learning_rate": 0.0002,
+      "loss": 0.3041,
+      "step": 6050
+    },
+    {
+      "epoch": 6.633825944170772,
+      "grad_norm": 1.3589898347854614,
+      "learning_rate": 0.0002,
+      "loss": 0.3075,
+      "step": 6060
+    },
+    {
+      "epoch": 6.644772851669403,
+      "grad_norm": 1.3488036394119263,
+      "learning_rate": 0.0002,
+      "loss": 0.3291,
+      "step": 6070
+    },
+    {
+      "epoch": 6.655719759168035,
+      "grad_norm": 1.0954102277755737,
+      "learning_rate": 0.0002,
+      "loss": 0.2898,
+      "step": 6080
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.4431062936782837,
+      "learning_rate": 0.0002,
+      "loss": 0.3489,
+      "step": 6090
+    },
+    {
+      "epoch": 6.677613574165298,
+      "grad_norm": 1.4387465715408325,
+      "learning_rate": 0.0002,
+      "loss": 0.2816,
+      "step": 6100
+    },
+    {
+      "epoch": 6.68856048166393,
+      "grad_norm": 1.8398990631103516,
+      "learning_rate": 0.0002,
+      "loss": 0.2705,
+      "step": 6110
+    },
+    {
+      "epoch": 6.699507389162561,
+      "grad_norm": 1.3523273468017578,
+      "learning_rate": 0.0002,
+      "loss": 0.3214,
+      "step": 6120
+    },
+    {
+      "epoch": 6.710454296661193,
+      "grad_norm": 1.6326191425323486,
+      "learning_rate": 0.0002,
+      "loss": 0.287,
+      "step": 6130
+    },
+    {
+      "epoch": 6.721401204159825,
+      "grad_norm": 1.3677960634231567,
+      "learning_rate": 0.0002,
+      "loss": 0.2857,
+      "step": 6140
+    },
+    {
+      "epoch": 6.732348111658457,
+      "grad_norm": 1.1993201971054077,
+      "learning_rate": 0.0002,
+      "loss": 0.3264,
+      "step": 6150
+    },
+    {
+      "epoch": 6.743295019157088,
+      "grad_norm": 1.1864078044891357,
+      "learning_rate": 0.0002,
+      "loss": 0.3071,
+      "step": 6160
+    },
+    {
+      "epoch": 6.75424192665572,
+      "grad_norm": 1.1625522375106812,
+      "learning_rate": 0.0002,
+      "loss": 0.3087,
+      "step": 6170
+    },
+    {
+      "epoch": 6.765188834154351,
+      "grad_norm": 1.5803234577178955,
+      "learning_rate": 0.0002,
+      "loss": 0.3551,
+      "step": 6180
+    },
+    {
+      "epoch": 6.776135741652983,
+      "grad_norm": 1.151746153831482,
+      "learning_rate": 0.0002,
+      "loss": 0.3059,
+      "step": 6190
+    },
+    {
+      "epoch": 6.787082649151615,
+      "grad_norm": 1.0727161169052124,
+      "learning_rate": 0.0002,
+      "loss": 0.2697,
+      "step": 6200
+    },
+    {
+      "epoch": 6.798029556650246,
+      "grad_norm": 1.4148162603378296,
+      "learning_rate": 0.0002,
+      "loss": 0.2844,
+      "step": 6210
+    },
+    {
+      "epoch": 6.808976464148878,
+      "grad_norm": 1.2071447372436523,
+      "learning_rate": 0.0002,
+      "loss": 0.3417,
+      "step": 6220
+    },
+    {
+      "epoch": 6.819923371647509,
+      "grad_norm": 1.3843804597854614,
+      "learning_rate": 0.0002,
+      "loss": 0.3066,
+      "step": 6230
+    },
+    {
+      "epoch": 6.830870279146141,
+      "grad_norm": 1.2490662336349487,
+      "learning_rate": 0.0002,
+      "loss": 0.2769,
+      "step": 6240
+    },
+    {
+      "epoch": 6.841817186644773,
+      "grad_norm": 1.6029689311981201,
+      "learning_rate": 0.0002,
+      "loss": 0.3237,
+      "step": 6250
+    },
+    {
+      "epoch": 6.852764094143405,
+      "grad_norm": 1.0388455390930176,
+      "learning_rate": 0.0002,
+      "loss": 0.3152,
+      "step": 6260
+    },
+    {
+      "epoch": 6.863711001642036,
+      "grad_norm": 1.3883857727050781,
+      "learning_rate": 0.0002,
+      "loss": 0.3026,
+      "step": 6270
+    },
+    {
+      "epoch": 6.874657909140668,
+      "grad_norm": 1.0500187873840332,
+      "learning_rate": 0.0002,
+      "loss": 0.3175,
+      "step": 6280
+    },
+    {
+      "epoch": 6.885604816639299,
+      "grad_norm": 1.4243487119674683,
+      "learning_rate": 0.0002,
+      "loss": 0.2952,
+      "step": 6290
+    },
+    {
+      "epoch": 6.896551724137931,
+      "grad_norm": 1.3169665336608887,
+      "learning_rate": 0.0002,
+      "loss": 0.2679,
+      "step": 6300
+    },
+    {
+      "epoch": 6.907498631636563,
+      "grad_norm": 1.5261493921279907,
+      "learning_rate": 0.0002,
+      "loss": 0.3291,
+      "step": 6310
+    },
+    {
+      "epoch": 6.9184455391351944,
+      "grad_norm": 1.578403115272522,
+      "learning_rate": 0.0002,
+      "loss": 0.3344,
+      "step": 6320
+    },
+    {
+      "epoch": 6.929392446633826,
+      "grad_norm": 1.4093263149261475,
+      "learning_rate": 0.0002,
+      "loss": 0.3263,
+      "step": 6330
+    },
+    {
+      "epoch": 6.940339354132457,
+      "grad_norm": 1.4003552198410034,
+      "learning_rate": 0.0002,
+      "loss": 0.3396,
+      "step": 6340
+    },
+    {
+      "epoch": 6.951286261631089,
+      "grad_norm": 1.650190830230713,
+      "learning_rate": 0.0002,
+      "loss": 0.3476,
+      "step": 6350
+    },
+    {
+      "epoch": 6.962233169129721,
+      "grad_norm": 1.2314515113830566,
+      "learning_rate": 0.0002,
+      "loss": 0.3442,
+      "step": 6360
+    },
+    {
+      "epoch": 6.973180076628353,
+      "grad_norm": 1.270980954170227,
+      "learning_rate": 0.0002,
+      "loss": 0.3341,
+      "step": 6370
+    },
+    {
+      "epoch": 6.984126984126984,
+      "grad_norm": 1.6352545022964478,
+      "learning_rate": 0.0002,
+      "loss": 0.3425,
+      "step": 6380
+    },
+    {
+      "epoch": 6.995073891625616,
+      "grad_norm": 1.3744925260543823,
+      "learning_rate": 0.0002,
+      "loss": 0.3647,
+      "step": 6390
+    },
+    {
+      "epoch": 6.999452654625069,
+      "eval_loss": 1.756764531135559,
+      "eval_runtime": 46.0542,
+      "eval_samples_per_second": 9.467,
+      "eval_steps_per_second": 1.194,
+      "step": 6394
+    },
+    {
+      "epoch": 7.006020799124247,
+      "grad_norm": 0.856991171836853,
+      "learning_rate": 0.0002,
+      "loss": 0.2356,
+      "step": 6400
+    },
+    {
+      "epoch": 7.016967706622879,
+      "grad_norm": 0.9483422040939331,
+      "learning_rate": 0.0002,
+      "loss": 0.2138,
+      "step": 6410
+    },
+    {
+      "epoch": 7.027914614121511,
+      "grad_norm": 1.0703433752059937,
+      "learning_rate": 0.0002,
+      "loss": 0.1884,
+      "step": 6420
+    },
+    {
+      "epoch": 7.0388615216201424,
+      "grad_norm": 1.761413812637329,
+      "learning_rate": 0.0002,
+      "loss": 0.2088,
+      "step": 6430
+    },
+    {
+      "epoch": 7.049808429118774,
+      "grad_norm": 0.9484238028526306,
+      "learning_rate": 0.0002,
+      "loss": 0.2156,
+      "step": 6440
+    },
+    {
+      "epoch": 7.060755336617405,
+      "grad_norm": 1.5663186311721802,
+      "learning_rate": 0.0002,
+      "loss": 0.2323,
+      "step": 6450
+    },
+    {
+      "epoch": 7.071702244116037,
+      "grad_norm": 0.7692174315452576,
+      "learning_rate": 0.0002,
+      "loss": 0.2089,
+      "step": 6460
+    },
+    {
+      "epoch": 7.082649151614669,
+      "grad_norm": 1.3554800748825073,
+      "learning_rate": 0.0002,
+      "loss": 0.1856,
+      "step": 6470
+    },
+    {
+      "epoch": 7.093596059113301,
+      "grad_norm": 0.9705919027328491,
+      "learning_rate": 0.0002,
+      "loss": 0.2057,
+      "step": 6480
+    },
+    {
+      "epoch": 7.104542966611932,
+      "grad_norm": 1.355778694152832,
+      "learning_rate": 0.0002,
+      "loss": 0.2068,
+      "step": 6490
+    },
+    {
+      "epoch": 7.115489874110564,
+      "grad_norm": 1.5389477014541626,
+      "learning_rate": 0.0002,
+      "loss": 0.2021,
+      "step": 6500
+    },
+    {
+      "epoch": 7.126436781609195,
+      "grad_norm": 0.9565434455871582,
+      "learning_rate": 0.0002,
+      "loss": 0.1963,
+      "step": 6510
+    },
+    {
+      "epoch": 7.137383689107827,
+      "grad_norm": 1.101539134979248,
+      "learning_rate": 0.0002,
+      "loss": 0.2323,
+      "step": 6520
+    },
+    {
+      "epoch": 7.148330596606459,
+      "grad_norm": 0.925153374671936,
+      "learning_rate": 0.0002,
+      "loss": 0.2024,
+      "step": 6530
+    },
+    {
+      "epoch": 7.1592775041050905,
+      "grad_norm": 1.1609078645706177,
+      "learning_rate": 0.0002,
+      "loss": 0.1955,
+      "step": 6540
+    },
+    {
+      "epoch": 7.170224411603722,
+      "grad_norm": 0.8908484578132629,
+      "learning_rate": 0.0002,
+      "loss": 0.2022,
+      "step": 6550
+    },
+    {
+      "epoch": 7.181171319102353,
+      "grad_norm": 0.9066158533096313,
+      "learning_rate": 0.0002,
+      "loss": 0.2378,
+      "step": 6560
+    },
+    {
+      "epoch": 7.192118226600985,
+      "grad_norm": 1.3601553440093994,
+      "learning_rate": 0.0002,
+      "loss": 0.1955,
+      "step": 6570
+    },
+    {
+      "epoch": 7.203065134099617,
+      "grad_norm": 1.0034444332122803,
+      "learning_rate": 0.0002,
+      "loss": 0.241,
+      "step": 6580
+    },
+    {
+      "epoch": 7.214012041598249,
+      "grad_norm": 1.608299970626831,
+      "learning_rate": 0.0002,
+      "loss": 0.2134,
+      "step": 6590
+    },
+    {
+      "epoch": 7.22495894909688,
+      "grad_norm": 1.2889668941497803,
+      "learning_rate": 0.0002,
+      "loss": 0.2089,
+      "step": 6600
+    },
+    {
+      "epoch": 7.235905856595512,
+      "grad_norm": 0.9896159768104553,
+      "learning_rate": 0.0002,
+      "loss": 0.2405,
+      "step": 6610
+    },
+    {
+      "epoch": 7.246852764094143,
+      "grad_norm": 1.408511996269226,
+      "learning_rate": 0.0002,
+      "loss": 0.2091,
+      "step": 6620
+    },
+    {
+      "epoch": 7.257799671592775,
+      "grad_norm": 1.0823664665222168,
+      "learning_rate": 0.0002,
+      "loss": 0.1958,
+      "step": 6630
+    },
+    {
+      "epoch": 7.268746579091407,
+      "grad_norm": 1.027026891708374,
+      "learning_rate": 0.0002,
+      "loss": 0.2117,
+      "step": 6640
+    },
+    {
+      "epoch": 7.2796934865900385,
+      "grad_norm": 1.0922648906707764,
+      "learning_rate": 0.0002,
+      "loss": 0.2319,
+      "step": 6650
+    },
+    {
+      "epoch": 7.29064039408867,
+      "grad_norm": 1.3361082077026367,
+      "learning_rate": 0.0002,
+      "loss": 0.2367,
+      "step": 6660
+    },
+    {
+      "epoch": 7.301587301587301,
+      "grad_norm": 1.9565683603286743,
+      "learning_rate": 0.0002,
+      "loss": 0.2299,
+      "step": 6670
+    },
+    {
+      "epoch": 7.312534209085933,
+      "grad_norm": 1.413672685623169,
+      "learning_rate": 0.0002,
+      "loss": 0.2248,
+      "step": 6680
+    },
+    {
+      "epoch": 7.323481116584565,
+      "grad_norm": 1.121842384338379,
+      "learning_rate": 0.0002,
+      "loss": 0.2306,
+      "step": 6690
+    },
+    {
+      "epoch": 7.334428024083197,
+      "grad_norm": 1.0622057914733887,
+      "learning_rate": 0.0002,
+      "loss": 0.2222,
+      "step": 6700
+    },
+    {
+      "epoch": 7.345374931581828,
+      "grad_norm": 1.280921459197998,
+      "learning_rate": 0.0002,
+      "loss": 0.2387,
+      "step": 6710
+    },
+    {
+      "epoch": 7.35632183908046,
+      "grad_norm": 1.5295953750610352,
+      "learning_rate": 0.0002,
+      "loss": 0.2428,
+      "step": 6720
+    },
+    {
+      "epoch": 7.367268746579091,
+      "grad_norm": 1.4289230108261108,
+      "learning_rate": 0.0002,
+      "loss": 0.2149,
+      "step": 6730
+    },
+    {
+      "epoch": 7.378215654077723,
+      "grad_norm": 1.535111665725708,
+      "learning_rate": 0.0002,
+      "loss": 0.2172,
+      "step": 6740
+    },
+    {
+      "epoch": 7.389162561576355,
+      "grad_norm": 1.777826189994812,
+      "learning_rate": 0.0002,
+      "loss": 0.2262,
+      "step": 6750
+    },
+    {
+      "epoch": 7.4001094690749865,
+      "grad_norm": 1.5058139562606812,
+      "learning_rate": 0.0002,
+      "loss": 0.2246,
+      "step": 6760
+    },
+    {
+      "epoch": 7.411056376573618,
+      "grad_norm": 0.9381663203239441,
+      "learning_rate": 0.0002,
+      "loss": 0.2323,
+      "step": 6770
+    },
+    {
+      "epoch": 7.422003284072249,
+      "grad_norm": 1.4739434719085693,
+      "learning_rate": 0.0002,
+      "loss": 0.2078,
+      "step": 6780
+    },
+    {
+      "epoch": 7.432950191570881,
+      "grad_norm": 1.8703559637069702,
+      "learning_rate": 0.0002,
+      "loss": 0.2493,
+      "step": 6790
+    },
+    {
+      "epoch": 7.443897099069513,
+      "grad_norm": 1.2242027521133423,
+      "learning_rate": 0.0002,
+      "loss": 0.2554,
+      "step": 6800
+    },
+    {
+      "epoch": 7.454844006568145,
+      "grad_norm": 1.3950374126434326,
+      "learning_rate": 0.0002,
+      "loss": 0.2258,
+      "step": 6810
+    },
+    {
+      "epoch": 7.465790914066776,
+      "grad_norm": 1.461701512336731,
+      "learning_rate": 0.0002,
+      "loss": 0.2365,
+      "step": 6820
+    },
+    {
+      "epoch": 7.476737821565408,
+      "grad_norm": 1.4460340738296509,
+      "learning_rate": 0.0002,
+      "loss": 0.2302,
+      "step": 6830
+    },
+    {
+      "epoch": 7.487684729064039,
+      "grad_norm": 1.0341510772705078,
+      "learning_rate": 0.0002,
+      "loss": 0.2294,
+      "step": 6840
+    },
+    {
+      "epoch": 7.498631636562671,
+      "grad_norm": 0.8885145783424377,
+      "learning_rate": 0.0002,
+      "loss": 0.2338,
+      "step": 6850
+    },
+    {
+      "epoch": 7.509578544061303,
+      "grad_norm": 2.4326062202453613,
+      "learning_rate": 0.0002,
+      "loss": 0.2424,
+      "step": 6860
+    },
+    {
+      "epoch": 7.5205254515599345,
+      "grad_norm": 1.1390372514724731,
+      "learning_rate": 0.0002,
+      "loss": 0.2352,
+      "step": 6870
+    },
+    {
+      "epoch": 7.531472359058566,
+      "grad_norm": 1.2346464395523071,
+      "learning_rate": 0.0002,
+      "loss": 0.2184,
+      "step": 6880
+    },
+    {
+      "epoch": 7.542419266557197,
+      "grad_norm": 1.6705836057662964,
+      "learning_rate": 0.0002,
+      "loss": 0.2389,
+      "step": 6890
+    },
+    {
+      "epoch": 7.553366174055829,
+      "grad_norm": 0.8130379319190979,
+      "learning_rate": 0.0002,
+      "loss": 0.2346,
+      "step": 6900
+    },
+    {
+      "epoch": 7.564313081554461,
+      "grad_norm": 1.2974088191986084,
+      "learning_rate": 0.0002,
+      "loss": 0.2165,
+      "step": 6910
+    },
+    {
+      "epoch": 7.575259989053093,
+      "grad_norm": 1.3465348482131958,
+      "learning_rate": 0.0002,
+      "loss": 0.2328,
+      "step": 6920
+    },
+    {
+      "epoch": 7.586206896551724,
+      "grad_norm": 1.245126724243164,
+      "learning_rate": 0.0002,
+      "loss": 0.2789,
+      "step": 6930
+    },
+    {
+      "epoch": 7.597153804050356,
+      "grad_norm": 1.3736917972564697,
+      "learning_rate": 0.0002,
+      "loss": 0.2446,
+      "step": 6940
+    },
+    {
+      "epoch": 7.608100711548987,
+      "grad_norm": 1.340989351272583,
+      "learning_rate": 0.0002,
+      "loss": 0.2483,
+      "step": 6950
+    },
+    {
+      "epoch": 7.619047619047619,
+      "grad_norm": 1.1082850694656372,
+      "learning_rate": 0.0002,
+      "loss": 0.3014,
+      "step": 6960
+    },
+    {
+      "epoch": 7.629994526546251,
+      "grad_norm": 1.3829188346862793,
+      "learning_rate": 0.0002,
+      "loss": 0.2319,
+      "step": 6970
+    },
+    {
+      "epoch": 7.6409414340448825,
+      "grad_norm": 1.5384989976882935,
+      "learning_rate": 0.0002,
+      "loss": 0.2299,
+      "step": 6980
+    },
+    {
+      "epoch": 7.651888341543514,
+      "grad_norm": 1.1061540842056274,
+      "learning_rate": 0.0002,
+      "loss": 0.2457,
+      "step": 6990
+    },
+    {
+      "epoch": 7.662835249042145,
+      "grad_norm": 1.2673815488815308,
+      "learning_rate": 0.0002,
+      "loss": 0.2216,
+      "step": 7000
+    },
+    {
+      "epoch": 7.673782156540777,
+      "grad_norm": 1.2290737628936768,
+      "learning_rate": 0.0002,
+      "loss": 0.2397,
+      "step": 7010
+    },
+    {
+      "epoch": 7.684729064039409,
+      "grad_norm": 1.4055291414260864,
+      "learning_rate": 0.0002,
+      "loss": 0.2112,
+      "step": 7020
+    },
+    {
+      "epoch": 7.695675971538041,
+      "grad_norm": 1.7786750793457031,
+      "learning_rate": 0.0002,
+      "loss": 0.2548,
+      "step": 7030
+    },
+    {
+      "epoch": 7.706622879036672,
+      "grad_norm": 1.454209566116333,
+      "learning_rate": 0.0002,
+      "loss": 0.2241,
+      "step": 7040
+    },
+    {
+      "epoch": 7.717569786535304,
+      "grad_norm": 1.3995633125305176,
+      "learning_rate": 0.0002,
+      "loss": 0.2461,
+      "step": 7050
+    },
+    {
+      "epoch": 7.728516694033935,
+      "grad_norm": 1.7514715194702148,
+      "learning_rate": 0.0002,
+      "loss": 0.2785,
+      "step": 7060
+    },
+    {
+      "epoch": 7.739463601532567,
+      "grad_norm": 1.5538004636764526,
+      "learning_rate": 0.0002,
+      "loss": 0.2335,
+      "step": 7070
+    },
+    {
+      "epoch": 7.750410509031199,
+      "grad_norm": 1.122506856918335,
+      "learning_rate": 0.0002,
+      "loss": 0.2245,
+      "step": 7080
+    },
+    {
+      "epoch": 7.7613574165298305,
+      "grad_norm": 1.2445831298828125,
+      "learning_rate": 0.0002,
+      "loss": 0.2473,
+      "step": 7090
+    },
+    {
+      "epoch": 7.772304324028462,
+      "grad_norm": 1.1478949785232544,
+      "learning_rate": 0.0002,
+      "loss": 0.2421,
+      "step": 7100
+    },
+    {
+      "epoch": 7.783251231527093,
+      "grad_norm": 1.4352518320083618,
+      "learning_rate": 0.0002,
+      "loss": 0.2346,
+      "step": 7110
+    },
+    {
+      "epoch": 7.794198139025725,
+      "grad_norm": 1.511096715927124,
+      "learning_rate": 0.0002,
+      "loss": 0.2351,
+      "step": 7120
+    },
+    {
+      "epoch": 7.805145046524357,
+      "grad_norm": 1.2296271324157715,
+      "learning_rate": 0.0002,
+      "loss": 0.2586,
+      "step": 7130
+    },
+    {
+      "epoch": 7.816091954022989,
+      "grad_norm": 1.7886443138122559,
+      "learning_rate": 0.0002,
+      "loss": 0.2453,
+      "step": 7140
+    },
+    {
+      "epoch": 7.82703886152162,
+      "grad_norm": 1.8886322975158691,
+      "learning_rate": 0.0002,
+      "loss": 0.2645,
+      "step": 7150
+    },
+    {
+      "epoch": 7.837985769020252,
+      "grad_norm": 1.3493725061416626,
+      "learning_rate": 0.0002,
+      "loss": 0.2658,
+      "step": 7160
+    },
+    {
+      "epoch": 7.848932676518883,
+      "grad_norm": 1.379209041595459,
+      "learning_rate": 0.0002,
+      "loss": 0.2787,
+      "step": 7170
+    },
+    {
+      "epoch": 7.859879584017515,
+      "grad_norm": 0.9374330043792725,
+      "learning_rate": 0.0002,
+      "loss": 0.248,
+      "step": 7180
+    },
+    {
+      "epoch": 7.870826491516147,
+      "grad_norm": 1.0391291379928589,
+      "learning_rate": 0.0002,
+      "loss": 0.2412,
+      "step": 7190
+    },
+    {
+      "epoch": 7.8817733990147785,
+      "grad_norm": 1.2710281610488892,
+      "learning_rate": 0.0002,
+      "loss": 0.2648,
+      "step": 7200
+    },
+    {
+      "epoch": 7.89272030651341,
+      "grad_norm": 1.6858662366867065,
+      "learning_rate": 0.0002,
+      "loss": 0.2511,
+      "step": 7210
+    },
+    {
+      "epoch": 7.903667214012041,
+      "grad_norm": 1.0925853252410889,
+      "learning_rate": 0.0002,
+      "loss": 0.2667,
+      "step": 7220
+    },
+    {
+      "epoch": 7.914614121510673,
+      "grad_norm": 1.7404073476791382,
+      "learning_rate": 0.0002,
+      "loss": 0.2756,
+      "step": 7230
+    },
+    {
+      "epoch": 7.925561029009305,
+      "grad_norm": 1.284067153930664,
+      "learning_rate": 0.0002,
+      "loss": 0.2486,
+      "step": 7240
+    },
+    {
+      "epoch": 7.936507936507937,
+      "grad_norm": 1.3801543712615967,
+      "learning_rate": 0.0002,
+      "loss": 0.2394,
+      "step": 7250
+    },
+    {
+      "epoch": 7.947454844006568,
+      "grad_norm": 1.4068974256515503,
+      "learning_rate": 0.0002,
+      "loss": 0.2761,
+      "step": 7260
+    },
+    {
+      "epoch": 7.9584017515052,
+      "grad_norm": 1.770037055015564,
+      "learning_rate": 0.0002,
+      "loss": 0.2455,
+      "step": 7270
+    },
+    {
+      "epoch": 7.969348659003831,
+      "grad_norm": 1.473775029182434,
+      "learning_rate": 0.0002,
+      "loss": 0.2863,
+      "step": 7280
+    },
+    {
+      "epoch": 7.980295566502463,
+      "grad_norm": 1.4878343343734741,
+      "learning_rate": 0.0002,
+      "loss": 0.2722,
+      "step": 7290
+    },
+    {
+      "epoch": 7.991242474001095,
+      "grad_norm": 1.2178987264633179,
+      "learning_rate": 0.0002,
+      "loss": 0.2921,
+      "step": 7300
+    },
+    {
+      "epoch": 7.995621237000547,
+      "eval_loss": 1.9267498254776,
+      "eval_runtime": 46.049,
+      "eval_samples_per_second": 9.468,
+      "eval_steps_per_second": 1.194,
+      "step": 7304
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 3.7516079806808064e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-7304/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..34b84c2d4cfdf217d0df26e8708c3bd34577ba32
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c63de276026ae040ebafb983d9df61b6c14a219ecb096cfd19bb9cbbdf5202b
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ebbf61940b0cd58679a1f0a656e7f180625f247b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50d27f47005f9bfcb7cc81d95f6db11766464cf433ebe0ec3b9bfd4cd2fa81de
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2e77fe7c65ab745e9d0c59f7bb780c34a5e85c99
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7262fed64284ee400ff89a52ad0dc0c5a3cc8afbab1a554324105e268b7058a8
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c1aded4707a46db5a73a88fa0f1af5b5ac2eaa8a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d68854151aa7f3535bf90a34c0de0f805fa3a448a25d2ac0eae91da60149361
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..410338bf7cb217e54fd7fba275e5771720427762
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/trainer_state.json
@@ -0,0 +1,678 @@
+{
+  "best_metric": 1.158464789390564,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913",
+  "epoch": 0.9994526546250684,
+  "eval_steps": 10,
+  "global_step": 913,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010946907498631636,
+      "grad_norm": 0.7611560821533203,
+      "learning_rate": 0.0002,
+      "loss": 2.4039,
+      "step": 10
+    },
+    {
+      "epoch": 0.021893814997263273,
+      "grad_norm": 0.4633193612098694,
+      "learning_rate": 0.0002,
+      "loss": 1.8818,
+      "step": 20
+    },
+    {
+      "epoch": 0.03284072249589491,
+      "grad_norm": 0.49326154589653015,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 30
+    },
+    {
+      "epoch": 0.043787629994526546,
+      "grad_norm": 0.3943138122558594,
+      "learning_rate": 0.0002,
+      "loss": 1.3859,
+      "step": 40
+    },
+    {
+      "epoch": 0.05473453749315818,
+      "grad_norm": 0.43292930722236633,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 50
+    },
+    {
+      "epoch": 0.06568144499178982,
+      "grad_norm": 0.3431817591190338,
+      "learning_rate": 0.0002,
+      "loss": 1.2427,
+      "step": 60
+    },
+    {
+      "epoch": 0.07662835249042145,
+      "grad_norm": 0.38011446595191956,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 70
+    },
+    {
+      "epoch": 0.08757525998905309,
+      "grad_norm": 0.7441071271896362,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 80
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 0.3359833061695099,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 90
+    },
+    {
+      "epoch": 0.10946907498631636,
+      "grad_norm": 0.3724392354488373,
+      "learning_rate": 0.0002,
+      "loss": 1.2774,
+      "step": 100
+    },
+    {
+      "epoch": 0.120415982484948,
+      "grad_norm": 0.40673762559890747,
+      "learning_rate": 0.0002,
+      "loss": 1.216,
+      "step": 110
+    },
+    {
+      "epoch": 0.13136288998357964,
+      "grad_norm": 0.40036800503730774,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 120
+    },
+    {
+      "epoch": 0.1423097974822113,
+      "grad_norm": 2.844191312789917,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 130
+    },
+    {
+      "epoch": 0.1532567049808429,
+      "grad_norm": 0.3104734420776367,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 140
+    },
+    {
+      "epoch": 0.16420361247947454,
+      "grad_norm": 0.3266797959804535,
+      "learning_rate": 0.0002,
+      "loss": 1.0498,
+      "step": 150
+    },
+    {
+      "epoch": 0.17515051997810618,
+      "grad_norm": 0.3079199194908142,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 160
+    },
+    {
+      "epoch": 0.18609742747673783,
+      "grad_norm": 0.3872479498386383,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 170
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 0.38654500246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 180
+    },
+    {
+      "epoch": 0.20799124247400108,
+      "grad_norm": 0.2913552522659302,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 190
+    },
+    {
+      "epoch": 0.21893814997263272,
+      "grad_norm": 0.2960572838783264,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 200
+    },
+    {
+      "epoch": 0.22988505747126436,
+      "grad_norm": 0.5175501108169556,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 210
+    },
+    {
+      "epoch": 0.240831964969896,
+      "grad_norm": 1.2921574115753174,
+      "learning_rate": 0.0002,
+      "loss": 1.1556,
+      "step": 220
+    },
+    {
+      "epoch": 0.25177887246852765,
+      "grad_norm": 0.2675512135028839,
+      "learning_rate": 0.0002,
+      "loss": 1.1488,
+      "step": 230
+    },
+    {
+      "epoch": 0.2627257799671593,
+      "grad_norm": 0.3956190049648285,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 240
+    },
+    {
+      "epoch": 0.27367268746579093,
+      "grad_norm": 0.6022581458091736,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 250
+    },
+    {
+      "epoch": 0.2846195949644226,
+      "grad_norm": 1.1949563026428223,
+      "learning_rate": 0.0002,
+      "loss": 1.1981,
+      "step": 260
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.31173548102378845,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 270
+    },
+    {
+      "epoch": 0.3065134099616858,
+      "grad_norm": 0.2808472812175751,
+      "learning_rate": 0.0002,
+      "loss": 1.06,
+      "step": 280
+    },
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 0.3042023777961731,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 290
+    },
+    {
+      "epoch": 0.3284072249589491,
+      "grad_norm": 0.39915043115615845,
+      "learning_rate": 0.0002,
+      "loss": 1.3147,
+      "step": 300
+    },
+    {
+      "epoch": 0.3393541324575807,
+      "grad_norm": 0.39118197560310364,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 310
+    },
+    {
+      "epoch": 0.35030103995621237,
+      "grad_norm": 0.355010986328125,
+      "learning_rate": 0.0002,
+      "loss": 1.1363,
+      "step": 320
+    },
+    {
+      "epoch": 0.361247947454844,
+      "grad_norm": 0.29734086990356445,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 330
+    },
+    {
+      "epoch": 0.37219485495347565,
+      "grad_norm": 0.346096009016037,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 340
+    },
+    {
+      "epoch": 0.3831417624521073,
+      "grad_norm": 0.4829643666744232,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 350
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.4726872742176056,
+      "learning_rate": 0.0002,
+      "loss": 1.2808,
+      "step": 360
+    },
+    {
+      "epoch": 0.4050355774493706,
+      "grad_norm": 0.3130153715610504,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 370
+    },
+    {
+      "epoch": 0.41598248494800216,
+      "grad_norm": 0.5123590230941772,
+      "learning_rate": 0.0002,
+      "loss": 1.1842,
+      "step": 380
+    },
+    {
+      "epoch": 0.4269293924466338,
+      "grad_norm": 0.3444574773311615,
+      "learning_rate": 0.0002,
+      "loss": 1.1539,
+      "step": 390
+    },
+    {
+      "epoch": 0.43787629994526545,
+      "grad_norm": 0.5302175283432007,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 400
+    },
+    {
+      "epoch": 0.4488232074438971,
+      "grad_norm": 0.2713572680950165,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 410
+    },
+    {
+      "epoch": 0.45977011494252873,
+      "grad_norm": 0.29530611634254456,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 420
+    },
+    {
+      "epoch": 0.47071702244116037,
+      "grad_norm": 0.27282455563545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1721,
+      "step": 430
+    },
+    {
+      "epoch": 0.481663929939792,
+      "grad_norm": 0.2647949755191803,
+      "learning_rate": 0.0002,
+      "loss": 1.1213,
+      "step": 440
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 0.35509347915649414,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 450
+    },
+    {
+      "epoch": 0.5035577449370553,
+      "grad_norm": 0.1959609091281891,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 460
+    },
+    {
+      "epoch": 0.5145046524356869,
+      "grad_norm": 0.40090155601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.1889,
+      "step": 470
+    },
+    {
+      "epoch": 0.5254515599343186,
+      "grad_norm": 0.3354604244232178,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 480
+    },
+    {
+      "epoch": 0.5363984674329502,
+      "grad_norm": 0.2758506238460541,
+      "learning_rate": 0.0002,
+      "loss": 1.1024,
+      "step": 490
+    },
+    {
+      "epoch": 0.5473453749315819,
+      "grad_norm": 0.3629051744937897,
+      "learning_rate": 0.0002,
+      "loss": 1.1108,
+      "step": 500
+    },
+    {
+      "epoch": 0.5582922824302134,
+      "grad_norm": 0.30802229046821594,
+      "learning_rate": 0.0002,
+      "loss": 1.2236,
+      "step": 510
+    },
+    {
+      "epoch": 0.5692391899288451,
+      "grad_norm": 0.3099463880062103,
+      "learning_rate": 0.0002,
+      "loss": 1.0424,
+      "step": 520
+    },
+    {
+      "epoch": 0.5801860974274767,
+      "grad_norm": 0.42299067974090576,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 530
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.5392252802848816,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 540
+    },
+    {
+      "epoch": 0.60207991242474,
+      "grad_norm": 0.34768250584602356,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 550
+    },
+    {
+      "epoch": 0.6130268199233716,
+      "grad_norm": 0.28490015864372253,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 560
+    },
+    {
+      "epoch": 0.6239737274220033,
+      "grad_norm": 0.34787994623184204,
+      "learning_rate": 0.0002,
+      "loss": 1.1111,
+      "step": 570
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 0.29058772325515747,
+      "learning_rate": 0.0002,
+      "loss": 1.0759,
+      "step": 580
+    },
+    {
+      "epoch": 0.6458675424192666,
+      "grad_norm": 0.4063778817653656,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 590
+    },
+    {
+      "epoch": 0.6568144499178982,
+      "grad_norm": 0.9244267344474792,
+      "learning_rate": 0.0002,
+      "loss": 1.1432,
+      "step": 600
+    },
+    {
+      "epoch": 0.6677613574165299,
+      "grad_norm": 0.27605190873146057,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 610
+    },
+    {
+      "epoch": 0.6787082649151615,
+      "grad_norm": 0.34346821904182434,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 620
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3093279302120209,
+      "learning_rate": 0.0002,
+      "loss": 1.2195,
+      "step": 630
+    },
+    {
+      "epoch": 0.7006020799124247,
+      "grad_norm": 1.0069009065628052,
+      "learning_rate": 0.0002,
+      "loss": 1.2461,
+      "step": 640
+    },
+    {
+      "epoch": 0.7115489874110563,
+      "grad_norm": 0.5049130916595459,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 650
+    },
+    {
+      "epoch": 0.722495894909688,
+      "grad_norm": 0.3748924732208252,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 660
+    },
+    {
+      "epoch": 0.7334428024083196,
+      "grad_norm": 0.2964959144592285,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 670
+    },
+    {
+      "epoch": 0.7443897099069513,
+      "grad_norm": 0.4599986970424652,
+      "learning_rate": 0.0002,
+      "loss": 1.2617,
+      "step": 680
+    },
+    {
+      "epoch": 0.7553366174055829,
+      "grad_norm": 0.27292951941490173,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 690
+    },
+    {
+      "epoch": 0.7662835249042146,
+      "grad_norm": 0.3123566806316376,
+      "learning_rate": 0.0002,
+      "loss": 1.2047,
+      "step": 700
+    },
+    {
+      "epoch": 0.7772304324028462,
+      "grad_norm": 0.28310710191726685,
+      "learning_rate": 0.0002,
+      "loss": 1.0021,
+      "step": 710
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.3279992341995239,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 720
+    },
+    {
+      "epoch": 0.7991242474001095,
+      "grad_norm": 0.28179168701171875,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 730
+    },
+    {
+      "epoch": 0.8100711548987412,
+      "grad_norm": 0.31492987275123596,
+      "learning_rate": 0.0002,
+      "loss": 1.0602,
+      "step": 740
+    },
+    {
+      "epoch": 0.8210180623973727,
+      "grad_norm": 0.41821011900901794,
+      "learning_rate": 0.0002,
+      "loss": 1.2518,
+      "step": 750
+    },
+    {
+      "epoch": 0.8319649698960043,
+      "grad_norm": 0.325235515832901,
+      "learning_rate": 0.0002,
+      "loss": 1.1612,
+      "step": 760
+    },
+    {
+      "epoch": 0.842911877394636,
+      "grad_norm": 0.5366070866584778,
+      "learning_rate": 0.0002,
+      "loss": 1.27,
+      "step": 770
+    },
+    {
+      "epoch": 0.8538587848932676,
+      "grad_norm": 0.32570579648017883,
+      "learning_rate": 0.0002,
+      "loss": 1.0921,
+      "step": 780
+    },
+    {
+      "epoch": 0.8648056923918993,
+      "grad_norm": 0.3642968237400055,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 790
+    },
+    {
+      "epoch": 0.8757525998905309,
+      "grad_norm": 0.29713448882102966,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 800
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.23599444329738617,
+      "learning_rate": 0.0002,
+      "loss": 1.0978,
+      "step": 810
+    },
+    {
+      "epoch": 0.8976464148877942,
+      "grad_norm": 0.31522464752197266,
+      "learning_rate": 0.0002,
+      "loss": 1.1867,
+      "step": 820
+    },
+    {
+      "epoch": 0.9085933223864259,
+      "grad_norm": 0.32754790782928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0208,
+      "step": 830
+    },
+    {
+      "epoch": 0.9195402298850575,
+      "grad_norm": 0.22741089761257172,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 840
+    },
+    {
+      "epoch": 0.9304871373836892,
+      "grad_norm": 0.3089679777622223,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 850
+    },
+    {
+      "epoch": 0.9414340448823207,
+      "grad_norm": 0.27440521121025085,
+      "learning_rate": 0.0002,
+      "loss": 1.0354,
+      "step": 860
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.3498363792896271,
+      "learning_rate": 0.0002,
+      "loss": 1.0417,
+      "step": 870
+    },
+    {
+      "epoch": 0.963327859879584,
+      "grad_norm": 0.47151854634284973,
+      "learning_rate": 0.0002,
+      "loss": 1.269,
+      "step": 880
+    },
+    {
+      "epoch": 0.9742747673782156,
+      "grad_norm": 0.24756591022014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1174,
+      "step": 890
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.2600938677787781,
+      "learning_rate": 0.0002,
+      "loss": 1.0622,
+      "step": 900
+    },
+    {
+      "epoch": 0.9961685823754789,
+      "grad_norm": 0.2934586703777313,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 910
+    },
+    {
+      "epoch": 0.9994526546250684,
+      "eval_loss": 1.158464789390564,
+      "eval_runtime": 46.0774,
+      "eval_samples_per_second": 9.462,
+      "eval_steps_per_second": 1.194,
+      "step": 913
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7304,
+  "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": 4.692078163132416e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/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_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..895406585c61ed7f5a6760976fc06a40a6fd8b3b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a158bca77bc6d58bfffd2acd2e9e641d481e1857fcbdd68fee875b74797abbc4
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_log.jsonl b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..7214f869187d01d42b242622ce668714cbe28764
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/training_log.jsonl
@@ -0,0 +1,13 @@
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 3908.758598089218, "total_accumulated_duration": 3908.758598089218, "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": 2.4021, "grad_norm": 0.8313556909561157, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8802, "grad_norm": 0.47623705863952637, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5893, "grad_norm": 0.5663179755210876, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3836, "grad_norm": 0.4509344696998596, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3075, "grad_norm": 0.4606502056121826, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2417, "grad_norm": 0.3602466881275177, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3164, "grad_norm": 0.4144728481769562, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1654, "grad_norm": 0.655055046081543, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1713, "grad_norm": 0.3738102614879608, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2749, "grad_norm": 0.37755176424980164, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.2094, "grad_norm": 0.4289834797382355, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2596, "grad_norm": 0.41391080617904663, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2378, "grad_norm": 0.3661900758743286, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.22, "grad_norm": 0.3107413053512573, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0469, "grad_norm": 0.35515156388282776, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3406173288822174, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2864, "grad_norm": 0.3639129102230072, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1742, "grad_norm": 0.36974409222602844, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.0633, "grad_norm": 0.29038330912590027, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0815, "grad_norm": 0.39991503953933716, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.1327, "grad_norm": 0.4746928811073303, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1591, "grad_norm": 0.3260287344455719, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1432, "grad_norm": 0.2829096019268036, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2742, "grad_norm": 0.3664100170135498, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1872, "grad_norm": 0.3916007876396179, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1883, "grad_norm": 0.2758415937423706, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1854, "grad_norm": 0.3045436143875122, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.0588, "grad_norm": 0.28762319684028625, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.172, "grad_norm": 0.30335670709609985, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3107, "grad_norm": 1.0805344581604004, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2439, "grad_norm": 0.932421088218689, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1394, "grad_norm": 0.3658805191516876, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1959, "grad_norm": 0.27649226784706116, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1926, "grad_norm": 0.34725308418273926, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1601, "grad_norm": 0.427664190530777, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2862, "grad_norm": 0.7244759202003479, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1521, "grad_norm": 0.30189067125320435, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1841, "grad_norm": 0.3003649413585663, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1567, "grad_norm": 0.33777716755867004, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1752, "grad_norm": 0.49972546100616455, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1113, "grad_norm": 0.2706405520439148, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1263, "grad_norm": 0.29881909489631653, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1701, "grad_norm": 0.27096378803253174, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1208, "grad_norm": 0.27717822790145874, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1653, "grad_norm": 0.37846508622169495, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1256, "grad_norm": 0.19916996359825134, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1836, "grad_norm": 0.385631799697876, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.1549, "grad_norm": 0.3161792457103729, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1032, "grad_norm": 0.3254718780517578, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1116, "grad_norm": 0.42534834146499634, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2216, "grad_norm": 0.29915255308151245, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0444, "grad_norm": 0.31264087557792664, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.2547, "grad_norm": 0.43361231684684753, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1655, "grad_norm": 0.3988197147846222, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1138, "grad_norm": 0.35821226239204407, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.1018, "grad_norm": 0.2939063310623169, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1155, "grad_norm": 0.33339911699295044, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0756, "grad_norm": 0.271671324968338, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.117, "grad_norm": 0.3313842713832855, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1381, "grad_norm": 1.0742870569229126, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0613, "grad_norm": 0.27820339798927307, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2129, "grad_norm": 0.3206658959388733, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2172, "grad_norm": 0.32589420676231384, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2451, "grad_norm": 0.28985416889190674, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0927, "grad_norm": 0.348069965839386, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1437, "grad_norm": 0.3746453523635864, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1051, "grad_norm": 0.30947357416152954, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2616, "grad_norm": 0.43517717719078064, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.0856, "grad_norm": 0.2904577851295471, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2039, "grad_norm": 0.31168660521507263, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0014, "grad_norm": 0.29549655318260193, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2264, "grad_norm": 0.3187733292579651, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.246, "grad_norm": 0.26287132501602173, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0605, "grad_norm": 0.32368725538253784, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2512, "grad_norm": 0.37818416953086853, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1609, "grad_norm": 0.3156316578388214, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.2619, "grad_norm": 0.4735032916069031, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0928, "grad_norm": 0.35200339555740356, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1014, "grad_norm": 0.2805202007293701, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1171, "grad_norm": 0.30746331810951233, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0937, "grad_norm": 0.23246702551841736, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1871, "grad_norm": 0.31561368703842163, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0224, "grad_norm": 0.4917743504047394, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9732, "grad_norm": 0.2250479906797409, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0664, "grad_norm": 0.3228110671043396, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0333, "grad_norm": 0.27888670563697815, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0386, "grad_norm": 0.32544803619384766, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.2681, "grad_norm": 0.4826430380344391, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1169, "grad_norm": 0.25451406836509705, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0649, "grad_norm": 0.3584475517272949, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.1037, "grad_norm": 0.31104281544685364, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 4858.700972080231, "total_accumulated_duration": 4858.700972080231, "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": 2.4047, "grad_norm": 0.789733350276947, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8779, "grad_norm": 0.47955045104026794, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5739, "grad_norm": 0.588397741317749, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.385, "grad_norm": 0.4773741662502289, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3023, "grad_norm": 0.4253212511539459, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.24, "grad_norm": 0.3454934060573578, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3092, "grad_norm": 0.40852540731430054, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1656, "grad_norm": 0.6172734498977661, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1724, "grad_norm": 0.3402147889137268, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2754, "grad_norm": 0.3680196702480316, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.2169, "grad_norm": 0.39625784754753113, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2626, "grad_norm": 0.3857010006904602, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2429, "grad_norm": 0.35471343994140625, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2243, "grad_norm": 0.3129616677761078, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0507, "grad_norm": 0.3068332076072693, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1645, "grad_norm": 0.31592103838920593, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2925, "grad_norm": 0.4139435887336731, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.178, "grad_norm": 0.4458570182323456, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.0656, "grad_norm": 0.29550260305404663, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0839, "grad_norm": 0.2915268540382385, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.1373, "grad_norm": 0.46415194869041443, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1627, "grad_norm": 0.3066752851009369, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.146, "grad_norm": 0.27449464797973633, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2801, "grad_norm": 0.35353365540504456, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1906, "grad_norm": 0.504181444644928, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1918, "grad_norm": 0.27208924293518066, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1866, "grad_norm": 0.294197678565979, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.0601, "grad_norm": 0.28713667392730713, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1749, "grad_norm": 0.3039015829563141, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.319, "grad_norm": 0.4140840470790863, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2412, "grad_norm": 0.4447316825389862, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1376, "grad_norm": 0.3732347786426544, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1932, "grad_norm": 0.3674798607826233, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.193, "grad_norm": 0.39562341570854187, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1608, "grad_norm": 0.3756610155105591, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.28, "grad_norm": 0.39729052782058716, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.15, "grad_norm": 0.30219244956970215, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1785, "grad_norm": 0.29158663749694824, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1554, "grad_norm": 0.33580854535102844, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1751, "grad_norm": 0.6002793312072754, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1116, "grad_norm": 0.27264222502708435, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.127, "grad_norm": 0.29089173674583435, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1732, "grad_norm": 0.26952189207077026, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1211, "grad_norm": 0.27022993564605713, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1639, "grad_norm": 0.3734654486179352, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1243, "grad_norm": 0.20133642852306366, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1842, "grad_norm": 0.39844706654548645, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.1548, "grad_norm": 0.302224338054657, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1029, "grad_norm": 0.28309789299964905, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.4141816794872284, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2211, "grad_norm": 0.2904048562049866, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0401, "grad_norm": 0.317196249961853, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.2524, "grad_norm": 0.42272260785102844, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1669, "grad_norm": 0.4033694863319397, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1153, "grad_norm": 0.3533627688884735, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.103, "grad_norm": 0.29309213161468506, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1116, "grad_norm": 0.3232715129852295, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0783, "grad_norm": 0.2727753221988678, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1156, "grad_norm": 0.3401601314544678, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1339, "grad_norm": 0.8788238167762756, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0589, "grad_norm": 0.27555420994758606, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2142, "grad_norm": 0.3252176344394684, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.224, "grad_norm": 0.3039418160915375, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2451, "grad_norm": 0.2943780720233917, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0915, "grad_norm": 0.33507248759269714, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1423, "grad_norm": 0.3620617091655731, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1067, "grad_norm": 0.3180168569087982, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2561, "grad_norm": 0.3375033438205719, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.0856, "grad_norm": 0.29707568883895874, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2042, "grad_norm": 0.3066735565662384, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 0.9978, "grad_norm": 0.2804313898086548, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2349, "grad_norm": 0.3300948143005371, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.2455, "grad_norm": 0.29191678762435913, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0589, "grad_norm": 0.30186912417411804, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2503, "grad_norm": 0.40812981128692627, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1592, "grad_norm": 0.3179655969142914, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.2694, "grad_norm": 0.5074726939201355, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0917, "grad_norm": 0.3446512520313263, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1031, "grad_norm": 0.2681335508823395, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1202, "grad_norm": 0.3224271237850189, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.24366702139377594, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1914, "grad_norm": 0.3292049467563629, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0203, "grad_norm": 0.3458293676376343, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9739, "grad_norm": 0.22741839289665222, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0687, "grad_norm": 0.6688203811645508, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0351, "grad_norm": 0.274652361869812, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0416, "grad_norm": 0.3694289028644562, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.2681, "grad_norm": 0.4757626950740814, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1171, "grad_norm": 0.23742415010929108, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0625, "grad_norm": 0.349394291639328, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0922, "grad_norm": 0.2974051833152771, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 4803.790252447128, "total_accumulated_duration": 4803.790252447128, "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": 2.4028, "grad_norm": 0.795521080493927, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8791, "grad_norm": 0.4686722159385681, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5831, "grad_norm": 0.5660933256149292, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3807, "grad_norm": 1.227329969406128, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3097, "grad_norm": 0.5239948630332947, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2397, "grad_norm": 0.3550681173801422, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3199, "grad_norm": 0.5845893621444702, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1702, "grad_norm": 0.6099892854690552, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1724, "grad_norm": 0.33250030875205994, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2806, "grad_norm": 0.5137067437171936, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.2112, "grad_norm": 0.40557053685188293, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2597, "grad_norm": 0.3856102526187897, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2382, "grad_norm": 0.3643055856227875, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2199, "grad_norm": 0.3158377408981323, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0512, "grad_norm": 0.31467798352241516, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1622, "grad_norm": 0.33398908376693726, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2859, "grad_norm": 0.38519906997680664, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1762, "grad_norm": 0.36663818359375, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.067, "grad_norm": 0.282659113407135, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.079, "grad_norm": 0.2818695604801178, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.1359, "grad_norm": 0.39851438999176025, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1538, "grad_norm": 0.31774890422821045, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1428, "grad_norm": 0.2777709364891052, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2741, "grad_norm": 0.3904445469379425, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.4349442720413208, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1855, "grad_norm": 0.2749781310558319, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1879, "grad_norm": 0.3258635699748993, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.0624, "grad_norm": 0.30227264761924744, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1725, "grad_norm": 0.3817561864852905, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3129, "grad_norm": 0.39917275309562683, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2445, "grad_norm": 0.42792218923568726, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1396, "grad_norm": 0.4719575345516205, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1918, "grad_norm": 0.2811720371246338, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.195, "grad_norm": 0.3931334912776947, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1617, "grad_norm": 0.4623820185661316, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2825, "grad_norm": 0.45300203561782837, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.151, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.18, "grad_norm": 0.3179274797439575, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1562, "grad_norm": 0.37702685594558716, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.174, "grad_norm": 0.5325851440429688, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1093, "grad_norm": 0.266667902469635, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1297, "grad_norm": 0.3148106634616852, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1729, "grad_norm": 0.26062336564064026, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1211, "grad_norm": 0.2696126103401184, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1657, "grad_norm": 0.3823620676994324, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1242, "grad_norm": 0.3216341435909271, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1854, "grad_norm": 0.39548951387405396, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.1529, "grad_norm": 0.3132643699645996, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.0981, "grad_norm": 0.29527464509010315, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1112, "grad_norm": 0.386337012052536, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2175, "grad_norm": 0.3230147957801819, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0429, "grad_norm": 0.3255913257598877, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.2527, "grad_norm": 0.42526349425315857, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1682, "grad_norm": 0.3711594045162201, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.114, "grad_norm": 0.3699112832546234, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.1013, "grad_norm": 0.29559260606765747, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1161, "grad_norm": 0.35342633724212646, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0747, "grad_norm": 0.27879607677459717, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1178, "grad_norm": 0.34217947721481323, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1317, "grad_norm": 0.9936075806617737, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0571, "grad_norm": 0.27797362208366394, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2133, "grad_norm": 0.3185802400112152, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2187, "grad_norm": 0.3263399302959442, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2469, "grad_norm": 0.3052294850349426, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0882, "grad_norm": 0.3395131826400757, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.35718032717704773, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1039, "grad_norm": 0.37328389286994934, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2621, "grad_norm": 0.3185192048549652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.0847, "grad_norm": 0.2795262932777405, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2064, "grad_norm": 0.29796719551086426, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 0.9993, "grad_norm": 0.3695313632488251, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2245, "grad_norm": 0.31174781918525696, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.2443, "grad_norm": 0.2640160620212555, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0605, "grad_norm": 0.31823453307151794, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2524, "grad_norm": 0.46503135561943054, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1609, "grad_norm": 0.3179810345172882, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.2593, "grad_norm": 0.4537338614463806, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0941, "grad_norm": 0.4113718867301941, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1023, "grad_norm": 0.2963421046733856, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1194, "grad_norm": 0.33587151765823364, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0984, "grad_norm": 0.23537468910217285, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1927, "grad_norm": 0.31468865275382996, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0221, "grad_norm": 0.336098313331604, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9768, "grad_norm": 0.2583458721637726, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0669, "grad_norm": 0.3173564076423645, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0359, "grad_norm": 0.27985909581184387, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0416, "grad_norm": 0.3379824459552765, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.2637, "grad_norm": 0.4532661736011505, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1143, "grad_norm": 0.24834831058979034, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0629, "grad_norm": 0.2924191951751709, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0929, "grad_norm": 0.29473212361335754, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 3111.739435195923, "total_accumulated_duration": 3111.739435195923, "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": 2.4064, "grad_norm": 0.8315525650978088, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8858, "grad_norm": 0.4768979549407959, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.598, "grad_norm": 0.4891189634799957, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3819, "grad_norm": 0.5654677152633667, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3084, "grad_norm": 0.4613725244998932, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2362, "grad_norm": 0.3586723208427429, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3189, "grad_norm": 0.4401360750198364, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1714, "grad_norm": 0.9771654605865479, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1746, "grad_norm": 0.34469878673553467, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2769, "grad_norm": 0.3707507848739624, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.2157, "grad_norm": 0.4154447615146637, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2554, "grad_norm": 0.7811278700828552, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2418, "grad_norm": 0.3964453339576721, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.221, "grad_norm": 0.32972392439842224, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0478, "grad_norm": 0.328593373298645, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1592, "grad_norm": 0.33038654923439026, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2844, "grad_norm": 0.4016525149345398, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.172, "grad_norm": 0.38213640451431274, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.0633, "grad_norm": 0.2944932281970978, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0803, "grad_norm": 0.2967860698699951, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.1353, "grad_norm": 0.6144160032272339, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1559, "grad_norm": 0.3172452747821808, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1414, "grad_norm": 0.29329076409339905, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2705, "grad_norm": 0.3799499273300171, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1869, "grad_norm": 0.3606453835964203, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1898, "grad_norm": 0.2818147540092468, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1907, "grad_norm": 0.43860194087028503, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.0607, "grad_norm": 0.2921052575111389, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1744, "grad_norm": 0.30841436982154846, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3094, "grad_norm": 0.6104546785354614, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.244, "grad_norm": 1.3919378519058228, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1422, "grad_norm": 0.4001498222351074, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1936, "grad_norm": 0.2854520380496979, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.2009, "grad_norm": 0.361251562833786, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1582, "grad_norm": 0.3750649690628052, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2793, "grad_norm": 0.40161052346229553, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1501, "grad_norm": 0.3158400058746338, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1825, "grad_norm": 0.30031272768974304, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1562, "grad_norm": 0.33606818318367004, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1765, "grad_norm": 0.49571487307548523, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1089, "grad_norm": 0.28612610697746277, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1275, "grad_norm": 0.29380208253860474, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.173, "grad_norm": 0.2754780054092407, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1198, "grad_norm": 0.7208526134490967, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1645, "grad_norm": 0.5081837773323059, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1279, "grad_norm": 0.20326904952526093, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1823, "grad_norm": 0.368084192276001, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.1567, "grad_norm": 0.3025706112384796, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.101, "grad_norm": 0.3122895658016205, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1089, "grad_norm": 0.4389672577381134, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2199, "grad_norm": 0.38763660192489624, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.042, "grad_norm": 0.32445019483566284, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.2558, "grad_norm": 0.40850868821144104, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1671, "grad_norm": 0.4462431073188782, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1108, "grad_norm": 0.3874809741973877, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.1014, "grad_norm": 0.3020818531513214, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1136, "grad_norm": 0.31303995847702026, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.076, "grad_norm": 0.26297080516815186, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1163, "grad_norm": 0.3534374535083771, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1344, "grad_norm": 0.6948974132537842, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0619, "grad_norm": 0.27238744497299194, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2142, "grad_norm": 0.3345783054828644, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2194, "grad_norm": 0.34639596939086914, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2466, "grad_norm": 0.28842809796333313, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0916, "grad_norm": 0.3722135126590729, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1421, "grad_norm": 0.36706414818763733, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1063, "grad_norm": 0.2937333881855011, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2621, "grad_norm": 0.3463886082172394, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.0862, "grad_norm": 0.2808016538619995, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2038, "grad_norm": 0.3184432089328766, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 0.9983, "grad_norm": 0.29998502135276794, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2272, "grad_norm": 0.32185453176498413, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.2452, "grad_norm": 0.27004799246788025, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0642, "grad_norm": 0.3320857882499695, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2558, "grad_norm": 0.37585026025772095, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.16, "grad_norm": 0.32140469551086426, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.2662, "grad_norm": 0.418083518743515, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0936, "grad_norm": 0.42758217453956604, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1041, "grad_norm": 0.27828508615493774, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.119, "grad_norm": 0.3195190131664276, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0979, "grad_norm": 0.2453049123287201, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1904, "grad_norm": 0.3614264726638794, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0221, "grad_norm": 0.3742782175540924, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9742, "grad_norm": 0.22376364469528198, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0657, "grad_norm": 0.4903332591056824, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0393, "grad_norm": 0.2924495339393616, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0412, "grad_norm": 0.3286125063896179, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.2697, "grad_norm": 0.45306113362312317, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1151, "grad_norm": 0.236796036362648, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0625, "grad_norm": 0.25988534092903137, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0953, "grad_norm": 0.2976899743080139, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 4324.058840036392, "total_accumulated_duration": 4324.058840036392, "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": 2.4025, "grad_norm": 0.8413946628570557, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8765, "grad_norm": 0.4706946611404419, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5774, "grad_norm": 0.592414140701294, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3784, "grad_norm": 0.5266873240470886, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3048, "grad_norm": 0.42714616656303406, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2416, "grad_norm": 0.33732205629348755, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3164, "grad_norm": 0.3919900059700012, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1686, "grad_norm": 0.6971392035484314, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1688, "grad_norm": 0.351304829120636, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2758, "grad_norm": 0.36268818378448486, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.2094, "grad_norm": 0.41704121232032776, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2561, "grad_norm": 0.3970317542552948, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2395, "grad_norm": 0.3852194845676422, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2199, "grad_norm": 0.30831000208854675, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0462, "grad_norm": 0.31121566891670227, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1639, "grad_norm": 0.34039783477783203, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2867, "grad_norm": 0.36704662442207336, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1745, "grad_norm": 0.36384010314941406, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.0667, "grad_norm": 0.31086239218711853, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0798, "grad_norm": 0.35405558347702026, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.1348, "grad_norm": 0.6467314958572388, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1532, "grad_norm": 0.32341739535331726, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1381, "grad_norm": 0.27438628673553467, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.272, "grad_norm": 0.37504860758781433, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1841, "grad_norm": 0.38969483971595764, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1885, "grad_norm": 0.27647948265075684, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1869, "grad_norm": 0.30249834060668945, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.058, "grad_norm": 1.0534214973449707, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1712, "grad_norm": 0.30905938148498535, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3124, "grad_norm": 0.8192272186279297, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2442, "grad_norm": 0.7817836999893188, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1415, "grad_norm": 0.36925482749938965, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1926, "grad_norm": 0.2638542652130127, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.2021, "grad_norm": 0.39350607991218567, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1669, "grad_norm": 0.40172526240348816, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2842, "grad_norm": 0.8559852838516235, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1495, "grad_norm": 0.3007528781890869, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1819, "grad_norm": 0.3273051083087921, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1567, "grad_norm": 0.35510483384132385, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1748, "grad_norm": 0.574571430683136, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.105, "grad_norm": 0.26277998089790344, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1307, "grad_norm": 0.2898577153682709, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.26525044441223145, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1192, "grad_norm": 0.26191475987434387, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1653, "grad_norm": 0.4236069619655609, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1258, "grad_norm": 0.21045465767383575, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.182, "grad_norm": 0.3767169415950775, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.1549, "grad_norm": 0.30658408999443054, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1038, "grad_norm": 0.2885505259037018, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1115, "grad_norm": 0.3905075192451477, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.219, "grad_norm": 0.3061444163322449, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0396, "grad_norm": 0.34073352813720703, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.2557, "grad_norm": 0.43636074662208557, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1711, "grad_norm": 0.48204460740089417, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1158, "grad_norm": 0.375756174325943, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.0986, "grad_norm": 0.2920822203159332, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1137, "grad_norm": 0.3567981421947479, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0756, "grad_norm": 0.28140270709991455, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1183, "grad_norm": 0.335514098405838, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1349, "grad_norm": 1.211463212966919, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0611, "grad_norm": 0.26986172795295715, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2121, "grad_norm": 0.3156491816043854, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2252, "grad_norm": 0.3593895435333252, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2449, "grad_norm": 0.29723769426345825, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0915, "grad_norm": 0.37067872285842896, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1409, "grad_norm": 0.4170680344104767, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1095, "grad_norm": 0.6047097444534302, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2664, "grad_norm": 0.3834376037120819, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.0855, "grad_norm": 0.2889885902404785, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.205, "grad_norm": 0.32053008675575256, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 0.9979, "grad_norm": 0.2932562828063965, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2304, "grad_norm": 0.6452378034591675, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.2512, "grad_norm": 0.32636845111846924, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0591, "grad_norm": 0.544005274772644, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2528, "grad_norm": 0.4918079972267151, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1599, "grad_norm": 0.3194459080696106, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.2657, "grad_norm": 0.5542969107627869, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0955, "grad_norm": 0.368278443813324, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1052, "grad_norm": 0.287670373916626, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1203, "grad_norm": 0.3100526034832001, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.1008, "grad_norm": 0.25668758153915405, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.3156385123729706, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0224, "grad_norm": 0.33307573199272156, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9724, "grad_norm": 0.22630105912685394, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0674, "grad_norm": 0.3279208242893219, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0391, "grad_norm": 0.5596337914466858, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0405, "grad_norm": 0.32892531156539917, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.2733, "grad_norm": 0.4221343696117401, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1131, "grad_norm": 0.2666907012462616, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0692, "grad_norm": 0.2804788053035736, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0977, "grad_norm": 0.3064863085746765, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 0.9994526546250684, "step": 913, "epoch_duration": 1366.6045315265656, "total_accumulated_duration": 1366.6045315265656, "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": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}]}
+{"epoch": 2.0, "step": 1827, "epoch_duration": 1362.2936389446259, "total_accumulated_duration": 2728.8981704711914, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-913", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}]}
+{"epoch": 2.9994526546250686, "step": 2740, "epoch_duration": 1360.6272654533386, "total_accumulated_duration": 4089.52543592453, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}]}
+{"epoch": 4.0, "step": 3654, "epoch_duration": 1363.5929102897644, "total_accumulated_duration": 5453.118346214294, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}, {"eval_loss": 1.1722838878631592, "eval_runtime": 46.0829, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.194, "epoch": 2.9994526546250686, "step": 2740}, {"loss": 0.6443, "grad_norm": 0.6384502053260803, "learning_rate": 0.0002, "epoch": 3.0103995621237, "step": 2750}, {"loss": 0.7123, "grad_norm": 0.9928722381591797, "learning_rate": 0.0002, "epoch": 3.0213464696223316, "step": 2760}, {"loss": 0.6045, "grad_norm": 0.7813051342964172, "learning_rate": 0.0002, "epoch": 3.0322933771209635, "step": 2770}, {"loss": 0.6042, "grad_norm": 1.0202556848526, "learning_rate": 0.0002, "epoch": 3.043240284619595, "step": 2780}, {"loss": 0.6356, "grad_norm": 0.7581062316894531, "learning_rate": 0.0002, "epoch": 3.0541871921182264, "step": 2790}, {"loss": 0.6349, "grad_norm": 0.6252710223197937, "learning_rate": 0.0002, "epoch": 3.0651340996168583, "step": 2800}, {"loss": 0.645, "grad_norm": 0.7738662958145142, "learning_rate": 0.0002, "epoch": 3.07608100711549, "step": 2810}, {"loss": 0.627, "grad_norm": 0.7381885051727295, "learning_rate": 0.0002, "epoch": 3.0870279146141213, "step": 2820}, {"loss": 0.6371, "grad_norm": 0.9197564721107483, "learning_rate": 0.0002, "epoch": 3.097974822112753, "step": 2830}, {"loss": 0.723, "grad_norm": 1.000976800918579, "learning_rate": 0.0002, "epoch": 3.1089217296113847, "step": 2840}, {"loss": 0.6631, "grad_norm": 0.7559131383895874, "learning_rate": 0.0002, "epoch": 3.1198686371100166, "step": 2850}, {"loss": 0.6252, "grad_norm": 0.7213780879974365, "learning_rate": 0.0002, "epoch": 3.130815544608648, "step": 2860}, {"loss": 0.6501, "grad_norm": 0.945939838886261, "learning_rate": 0.0002, "epoch": 3.1417624521072796, "step": 2870}, {"loss": 0.6129, "grad_norm": 0.7277454137802124, "learning_rate": 0.0002, "epoch": 3.1527093596059115, "step": 2880}, {"loss": 0.6423, "grad_norm": 0.762026846408844, "learning_rate": 0.0002, "epoch": 3.163656267104543, "step": 2890}, {"loss": 0.5332, "grad_norm": 0.6471221446990967, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 2900}, {"loss": 0.7981, "grad_norm": 0.6018978357315063, "learning_rate": 0.0002, "epoch": 3.1855500821018063, "step": 2910}, {"loss": 0.7274, "grad_norm": 0.8607320785522461, "learning_rate": 0.0002, "epoch": 3.196496989600438, "step": 2920}, {"loss": 0.6139, "grad_norm": 0.8854126334190369, "learning_rate": 0.0002, "epoch": 3.2074438970990693, "step": 2930}, {"loss": 0.6485, "grad_norm": 0.6620870232582092, "learning_rate": 0.0002, "epoch": 3.218390804597701, "step": 2940}, {"loss": 0.6969, "grad_norm": 0.7377511858940125, "learning_rate": 0.0002, "epoch": 3.2293377120963327, "step": 2950}, {"loss": 0.6798, "grad_norm": 0.7803301811218262, "learning_rate": 0.0002, "epoch": 3.2402846195949646, "step": 2960}, {"loss": 0.6697, "grad_norm": 0.834061861038208, "learning_rate": 0.0002, "epoch": 3.251231527093596, "step": 2970}, {"loss": 0.6894, "grad_norm": 0.8496041893959045, "learning_rate": 0.0002, "epoch": 3.2621784345922276, "step": 2980}, {"loss": 0.6591, "grad_norm": 0.7967984676361084, "learning_rate": 0.0002, "epoch": 3.2731253420908595, "step": 2990}, {"loss": 0.7266, "grad_norm": 1.0207016468048096, "learning_rate": 0.0002, "epoch": 3.284072249589491, "step": 3000}, {"loss": 0.6586, "grad_norm": 0.7855771780014038, "learning_rate": 0.0002, "epoch": 3.2950191570881224, "step": 3010}, {"loss": 0.5711, "grad_norm": 0.9427546858787537, "learning_rate": 0.0002, "epoch": 3.3059660645867543, "step": 3020}, {"loss": 0.6277, "grad_norm": 0.823542594909668, "learning_rate": 0.0002, "epoch": 3.316912972085386, "step": 3030}, {"loss": 0.7109, "grad_norm": 0.9826635122299194, "learning_rate": 0.0002, "epoch": 3.3278598795840173, "step": 3040}, {"loss": 0.6564, "grad_norm": 0.7259827852249146, "learning_rate": 0.0002, "epoch": 3.338806787082649, "step": 3050}, {"loss": 0.653, "grad_norm": 0.7774739861488342, "learning_rate": 0.0002, "epoch": 3.3497536945812807, "step": 3060}, {"loss": 0.7529, "grad_norm": 0.7394293546676636, "learning_rate": 0.0002, "epoch": 3.3607006020799126, "step": 3070}, {"loss": 0.5987, "grad_norm": 0.9017578959465027, "learning_rate": 0.0002, "epoch": 3.371647509578544, "step": 3080}, {"loss": 0.6953, "grad_norm": 0.7451054453849792, "learning_rate": 0.0002, "epoch": 3.3825944170771756, "step": 3090}, {"loss": 0.6759, "grad_norm": 0.7321506142616272, "learning_rate": 0.0002, "epoch": 3.3935413245758075, "step": 3100}, {"loss": 0.6555, "grad_norm": 0.6721828579902649, "learning_rate": 0.0002, "epoch": 3.404488232074439, "step": 3110}, {"loss": 0.6559, "grad_norm": 0.774022102355957, "learning_rate": 0.0002, "epoch": 3.4154351395730704, "step": 3120}, {"loss": 0.7449, "grad_norm": 0.9143537282943726, "learning_rate": 0.0002, "epoch": 3.4263820470717024, "step": 3130}, {"loss": 0.6899, "grad_norm": 1.226087212562561, "learning_rate": 0.0002, "epoch": 3.437328954570334, "step": 3140}, {"loss": 0.6719, "grad_norm": 0.7545496225357056, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 3150}, {"loss": 0.6153, "grad_norm": 0.6515635848045349, "learning_rate": 0.0002, "epoch": 3.4592227695675972, "step": 3160}, {"loss": 0.6926, "grad_norm": 0.9297090172767639, "learning_rate": 0.0002, "epoch": 3.4701696770662287, "step": 3170}, {"loss": 0.6071, "grad_norm": 1.0130730867385864, "learning_rate": 0.0002, "epoch": 3.4811165845648606, "step": 3180}, {"loss": 0.5959, "grad_norm": 0.7654589414596558, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 3190}, {"loss": 0.7401, "grad_norm": 0.9954977631568909, "learning_rate": 0.0002, "epoch": 3.5030103995621236, "step": 3200}, {"loss": 0.6661, "grad_norm": 0.6027487516403198, "learning_rate": 0.0002, "epoch": 3.5139573070607555, "step": 3210}, {"loss": 0.6963, "grad_norm": 0.741770327091217, "learning_rate": 0.0002, "epoch": 3.524904214559387, "step": 3220}, {"loss": 0.8112, "grad_norm": 1.0534909963607788, "learning_rate": 0.0002, "epoch": 3.535851122058019, "step": 3230}, {"loss": 0.6813, "grad_norm": 0.937772274017334, "learning_rate": 0.0002, "epoch": 3.5467980295566504, "step": 3240}, {"loss": 0.6681, "grad_norm": 0.8504213690757751, "learning_rate": 0.0002, "epoch": 3.557744937055282, "step": 3250}, {"loss": 0.6436, "grad_norm": 0.7755007147789001, "learning_rate": 0.0002, "epoch": 3.5686918445539133, "step": 3260}, {"loss": 0.6213, "grad_norm": 1.0193358659744263, "learning_rate": 0.0002, "epoch": 3.5796387520525452, "step": 3270}, {"loss": 0.671, "grad_norm": 0.8440536856651306, "learning_rate": 0.0002, "epoch": 3.5905856595511767, "step": 3280}, {"loss": 0.6859, "grad_norm": 0.6195939183235168, "learning_rate": 0.0002, "epoch": 3.6015325670498086, "step": 3290}, {"loss": 0.7446, "grad_norm": 0.8608590960502625, "learning_rate": 0.0002, "epoch": 3.61247947454844, "step": 3300}, {"loss": 0.7301, "grad_norm": 0.6772327423095703, "learning_rate": 0.0002, "epoch": 3.6234263820470716, "step": 3310}, {"loss": 0.6298, "grad_norm": 0.8031839728355408, "learning_rate": 0.0002, "epoch": 3.6343732895457035, "step": 3320}, {"loss": 0.7041, "grad_norm": 0.6080502271652222, "learning_rate": 0.0002, "epoch": 3.645320197044335, "step": 3330}, {"loss": 0.7431, "grad_norm": 0.8007240891456604, "learning_rate": 0.0002, "epoch": 3.656267104542967, "step": 3340}, {"loss": 0.7446, "grad_norm": 0.8060704469680786, "learning_rate": 0.0002, "epoch": 3.6672140120415984, "step": 3350}, {"loss": 0.6304, "grad_norm": 0.7547586560249329, "learning_rate": 0.0002, "epoch": 3.67816091954023, "step": 3360}, {"loss": 0.7066, "grad_norm": 0.686851978302002, "learning_rate": 0.0002, "epoch": 3.6891078270388613, "step": 3370}, {"loss": 0.6748, "grad_norm": 0.9429075717926025, "learning_rate": 0.0002, "epoch": 3.7000547345374932, "step": 3380}, {"loss": 0.6673, "grad_norm": 0.7283591032028198, "learning_rate": 0.0002, "epoch": 3.7110016420361247, "step": 3390}, {"loss": 0.7502, "grad_norm": 0.8323085904121399, "learning_rate": 0.0002, "epoch": 3.7219485495347566, "step": 3400}, {"loss": 0.7779, "grad_norm": 0.8529590964317322, "learning_rate": 0.0002, "epoch": 3.732895457033388, "step": 3410}, {"loss": 0.6555, "grad_norm": 0.731752872467041, "learning_rate": 0.0002, "epoch": 3.7438423645320196, "step": 3420}, {"loss": 0.6928, "grad_norm": 0.8572278618812561, "learning_rate": 0.0002, "epoch": 3.7547892720306515, "step": 3430}, {"loss": 0.6215, "grad_norm": 0.7408691048622131, "learning_rate": 0.0002, "epoch": 3.765736179529283, "step": 3440}, {"loss": 0.622, "grad_norm": 0.7470445036888123, "learning_rate": 0.0002, "epoch": 3.776683087027915, "step": 3450}, {"loss": 0.7241, "grad_norm": 0.6806244254112244, "learning_rate": 0.0002, "epoch": 3.7876299945265464, "step": 3460}, {"loss": 0.7739, "grad_norm": 0.9129069447517395, "learning_rate": 0.0002, "epoch": 3.798576902025178, "step": 3470}, {"loss": 0.6826, "grad_norm": 0.8717501759529114, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 3480}, {"loss": 0.6188, "grad_norm": 0.6761979460716248, "learning_rate": 0.0002, "epoch": 3.8204707170224412, "step": 3490}, {"loss": 0.7601, "grad_norm": 1.0054380893707275, "learning_rate": 0.0002, "epoch": 3.8314176245210727, "step": 3500}, {"loss": 0.623, "grad_norm": 1.1224009990692139, "learning_rate": 0.0002, "epoch": 3.8423645320197046, "step": 3510}, {"loss": 0.6918, "grad_norm": 0.8997692465782166, "learning_rate": 0.0002, "epoch": 3.853311439518336, "step": 3520}, {"loss": 0.6357, "grad_norm": 1.0086902379989624, "learning_rate": 0.0002, "epoch": 3.8642583470169676, "step": 3530}, {"loss": 0.6379, "grad_norm": 0.772739589214325, "learning_rate": 0.0002, "epoch": 3.8752052545155995, "step": 3540}, {"loss": 0.7423, "grad_norm": 1.211774230003357, "learning_rate": 0.0002, "epoch": 3.886152162014231, "step": 3550}, {"loss": 0.7321, "grad_norm": 0.9572356939315796, "learning_rate": 0.0002, "epoch": 3.897099069512863, "step": 3560}, {"loss": 0.6836, "grad_norm": 0.7887842655181885, "learning_rate": 0.0002, "epoch": 3.9080459770114944, "step": 3570}, {"loss": 0.7576, "grad_norm": 0.7308389544487, "learning_rate": 0.0002, "epoch": 3.918992884510126, "step": 3580}, {"loss": 0.6001, "grad_norm": 1.0182650089263916, "learning_rate": 0.0002, "epoch": 3.9299397920087573, "step": 3590}, {"loss": 0.6942, "grad_norm": 0.8000147342681885, "learning_rate": 0.0002, "epoch": 3.9408866995073892, "step": 3600}, {"loss": 0.6244, "grad_norm": 0.7385728359222412, "learning_rate": 0.0002, "epoch": 3.9518336070060207, "step": 3610}, {"loss": 0.6718, "grad_norm": 0.9233261942863464, "learning_rate": 0.0002, "epoch": 3.9627805145046526, "step": 3620}, {"loss": 0.6508, "grad_norm": 0.8486751914024353, "learning_rate": 0.0002, "epoch": 3.973727422003284, "step": 3630}, {"loss": 0.6928, "grad_norm": 0.7593663334846497, "learning_rate": 0.0002, "epoch": 3.9846743295019156, "step": 3640}, {"loss": 0.6847, "grad_norm": 0.7885415554046631, "learning_rate": 0.0002, "epoch": 3.9956212370005475, "step": 3650}]}
+{"epoch": 4.999452654625069, "step": 4567, "epoch_duration": 1358.2625460624695, "total_accumulated_duration": 6811.380892276764, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}, {"eval_loss": 1.1722838878631592, "eval_runtime": 46.0829, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.194, "epoch": 2.9994526546250686, "step": 2740}, {"loss": 0.6443, "grad_norm": 0.6384502053260803, "learning_rate": 0.0002, "epoch": 3.0103995621237, "step": 2750}, {"loss": 0.7123, "grad_norm": 0.9928722381591797, "learning_rate": 0.0002, "epoch": 3.0213464696223316, "step": 2760}, {"loss": 0.6045, "grad_norm": 0.7813051342964172, "learning_rate": 0.0002, "epoch": 3.0322933771209635, "step": 2770}, {"loss": 0.6042, "grad_norm": 1.0202556848526, "learning_rate": 0.0002, "epoch": 3.043240284619595, "step": 2780}, {"loss": 0.6356, "grad_norm": 0.7581062316894531, "learning_rate": 0.0002, "epoch": 3.0541871921182264, "step": 2790}, {"loss": 0.6349, "grad_norm": 0.6252710223197937, "learning_rate": 0.0002, "epoch": 3.0651340996168583, "step": 2800}, {"loss": 0.645, "grad_norm": 0.7738662958145142, "learning_rate": 0.0002, "epoch": 3.07608100711549, "step": 2810}, {"loss": 0.627, "grad_norm": 0.7381885051727295, "learning_rate": 0.0002, "epoch": 3.0870279146141213, "step": 2820}, {"loss": 0.6371, "grad_norm": 0.9197564721107483, "learning_rate": 0.0002, "epoch": 3.097974822112753, "step": 2830}, {"loss": 0.723, "grad_norm": 1.000976800918579, "learning_rate": 0.0002, "epoch": 3.1089217296113847, "step": 2840}, {"loss": 0.6631, "grad_norm": 0.7559131383895874, "learning_rate": 0.0002, "epoch": 3.1198686371100166, "step": 2850}, {"loss": 0.6252, "grad_norm": 0.7213780879974365, "learning_rate": 0.0002, "epoch": 3.130815544608648, "step": 2860}, {"loss": 0.6501, "grad_norm": 0.945939838886261, "learning_rate": 0.0002, "epoch": 3.1417624521072796, "step": 2870}, {"loss": 0.6129, "grad_norm": 0.7277454137802124, "learning_rate": 0.0002, "epoch": 3.1527093596059115, "step": 2880}, {"loss": 0.6423, "grad_norm": 0.762026846408844, "learning_rate": 0.0002, "epoch": 3.163656267104543, "step": 2890}, {"loss": 0.5332, "grad_norm": 0.6471221446990967, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 2900}, {"loss": 0.7981, "grad_norm": 0.6018978357315063, "learning_rate": 0.0002, "epoch": 3.1855500821018063, "step": 2910}, {"loss": 0.7274, "grad_norm": 0.8607320785522461, "learning_rate": 0.0002, "epoch": 3.196496989600438, "step": 2920}, {"loss": 0.6139, "grad_norm": 0.8854126334190369, "learning_rate": 0.0002, "epoch": 3.2074438970990693, "step": 2930}, {"loss": 0.6485, "grad_norm": 0.6620870232582092, "learning_rate": 0.0002, "epoch": 3.218390804597701, "step": 2940}, {"loss": 0.6969, "grad_norm": 0.7377511858940125, "learning_rate": 0.0002, "epoch": 3.2293377120963327, "step": 2950}, {"loss": 0.6798, "grad_norm": 0.7803301811218262, "learning_rate": 0.0002, "epoch": 3.2402846195949646, "step": 2960}, {"loss": 0.6697, "grad_norm": 0.834061861038208, "learning_rate": 0.0002, "epoch": 3.251231527093596, "step": 2970}, {"loss": 0.6894, "grad_norm": 0.8496041893959045, "learning_rate": 0.0002, "epoch": 3.2621784345922276, "step": 2980}, {"loss": 0.6591, "grad_norm": 0.7967984676361084, "learning_rate": 0.0002, "epoch": 3.2731253420908595, "step": 2990}, {"loss": 0.7266, "grad_norm": 1.0207016468048096, "learning_rate": 0.0002, "epoch": 3.284072249589491, "step": 3000}, {"loss": 0.6586, "grad_norm": 0.7855771780014038, "learning_rate": 0.0002, "epoch": 3.2950191570881224, "step": 3010}, {"loss": 0.5711, "grad_norm": 0.9427546858787537, "learning_rate": 0.0002, "epoch": 3.3059660645867543, "step": 3020}, {"loss": 0.6277, "grad_norm": 0.823542594909668, "learning_rate": 0.0002, "epoch": 3.316912972085386, "step": 3030}, {"loss": 0.7109, "grad_norm": 0.9826635122299194, "learning_rate": 0.0002, "epoch": 3.3278598795840173, "step": 3040}, {"loss": 0.6564, "grad_norm": 0.7259827852249146, "learning_rate": 0.0002, "epoch": 3.338806787082649, "step": 3050}, {"loss": 0.653, "grad_norm": 0.7774739861488342, "learning_rate": 0.0002, "epoch": 3.3497536945812807, "step": 3060}, {"loss": 0.7529, "grad_norm": 0.7394293546676636, "learning_rate": 0.0002, "epoch": 3.3607006020799126, "step": 3070}, {"loss": 0.5987, "grad_norm": 0.9017578959465027, "learning_rate": 0.0002, "epoch": 3.371647509578544, "step": 3080}, {"loss": 0.6953, "grad_norm": 0.7451054453849792, "learning_rate": 0.0002, "epoch": 3.3825944170771756, "step": 3090}, {"loss": 0.6759, "grad_norm": 0.7321506142616272, "learning_rate": 0.0002, "epoch": 3.3935413245758075, "step": 3100}, {"loss": 0.6555, "grad_norm": 0.6721828579902649, "learning_rate": 0.0002, "epoch": 3.404488232074439, "step": 3110}, {"loss": 0.6559, "grad_norm": 0.774022102355957, "learning_rate": 0.0002, "epoch": 3.4154351395730704, "step": 3120}, {"loss": 0.7449, "grad_norm": 0.9143537282943726, "learning_rate": 0.0002, "epoch": 3.4263820470717024, "step": 3130}, {"loss": 0.6899, "grad_norm": 1.226087212562561, "learning_rate": 0.0002, "epoch": 3.437328954570334, "step": 3140}, {"loss": 0.6719, "grad_norm": 0.7545496225357056, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 3150}, {"loss": 0.6153, "grad_norm": 0.6515635848045349, "learning_rate": 0.0002, "epoch": 3.4592227695675972, "step": 3160}, {"loss": 0.6926, "grad_norm": 0.9297090172767639, "learning_rate": 0.0002, "epoch": 3.4701696770662287, "step": 3170}, {"loss": 0.6071, "grad_norm": 1.0130730867385864, "learning_rate": 0.0002, "epoch": 3.4811165845648606, "step": 3180}, {"loss": 0.5959, "grad_norm": 0.7654589414596558, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 3190}, {"loss": 0.7401, "grad_norm": 0.9954977631568909, "learning_rate": 0.0002, "epoch": 3.5030103995621236, "step": 3200}, {"loss": 0.6661, "grad_norm": 0.6027487516403198, "learning_rate": 0.0002, "epoch": 3.5139573070607555, "step": 3210}, {"loss": 0.6963, "grad_norm": 0.741770327091217, "learning_rate": 0.0002, "epoch": 3.524904214559387, "step": 3220}, {"loss": 0.8112, "grad_norm": 1.0534909963607788, "learning_rate": 0.0002, "epoch": 3.535851122058019, "step": 3230}, {"loss": 0.6813, "grad_norm": 0.937772274017334, "learning_rate": 0.0002, "epoch": 3.5467980295566504, "step": 3240}, {"loss": 0.6681, "grad_norm": 0.8504213690757751, "learning_rate": 0.0002, "epoch": 3.557744937055282, "step": 3250}, {"loss": 0.6436, "grad_norm": 0.7755007147789001, "learning_rate": 0.0002, "epoch": 3.5686918445539133, "step": 3260}, {"loss": 0.6213, "grad_norm": 1.0193358659744263, "learning_rate": 0.0002, "epoch": 3.5796387520525452, "step": 3270}, {"loss": 0.671, "grad_norm": 0.8440536856651306, "learning_rate": 0.0002, "epoch": 3.5905856595511767, "step": 3280}, {"loss": 0.6859, "grad_norm": 0.6195939183235168, "learning_rate": 0.0002, "epoch": 3.6015325670498086, "step": 3290}, {"loss": 0.7446, "grad_norm": 0.8608590960502625, "learning_rate": 0.0002, "epoch": 3.61247947454844, "step": 3300}, {"loss": 0.7301, "grad_norm": 0.6772327423095703, "learning_rate": 0.0002, "epoch": 3.6234263820470716, "step": 3310}, {"loss": 0.6298, "grad_norm": 0.8031839728355408, "learning_rate": 0.0002, "epoch": 3.6343732895457035, "step": 3320}, {"loss": 0.7041, "grad_norm": 0.6080502271652222, "learning_rate": 0.0002, "epoch": 3.645320197044335, "step": 3330}, {"loss": 0.7431, "grad_norm": 0.8007240891456604, "learning_rate": 0.0002, "epoch": 3.656267104542967, "step": 3340}, {"loss": 0.7446, "grad_norm": 0.8060704469680786, "learning_rate": 0.0002, "epoch": 3.6672140120415984, "step": 3350}, {"loss": 0.6304, "grad_norm": 0.7547586560249329, "learning_rate": 0.0002, "epoch": 3.67816091954023, "step": 3360}, {"loss": 0.7066, "grad_norm": 0.686851978302002, "learning_rate": 0.0002, "epoch": 3.6891078270388613, "step": 3370}, {"loss": 0.6748, "grad_norm": 0.9429075717926025, "learning_rate": 0.0002, "epoch": 3.7000547345374932, "step": 3380}, {"loss": 0.6673, "grad_norm": 0.7283591032028198, "learning_rate": 0.0002, "epoch": 3.7110016420361247, "step": 3390}, {"loss": 0.7502, "grad_norm": 0.8323085904121399, "learning_rate": 0.0002, "epoch": 3.7219485495347566, "step": 3400}, {"loss": 0.7779, "grad_norm": 0.8529590964317322, "learning_rate": 0.0002, "epoch": 3.732895457033388, "step": 3410}, {"loss": 0.6555, "grad_norm": 0.731752872467041, "learning_rate": 0.0002, "epoch": 3.7438423645320196, "step": 3420}, {"loss": 0.6928, "grad_norm": 0.8572278618812561, "learning_rate": 0.0002, "epoch": 3.7547892720306515, "step": 3430}, {"loss": 0.6215, "grad_norm": 0.7408691048622131, "learning_rate": 0.0002, "epoch": 3.765736179529283, "step": 3440}, {"loss": 0.622, "grad_norm": 0.7470445036888123, "learning_rate": 0.0002, "epoch": 3.776683087027915, "step": 3450}, {"loss": 0.7241, "grad_norm": 0.6806244254112244, "learning_rate": 0.0002, "epoch": 3.7876299945265464, "step": 3460}, {"loss": 0.7739, "grad_norm": 0.9129069447517395, "learning_rate": 0.0002, "epoch": 3.798576902025178, "step": 3470}, {"loss": 0.6826, "grad_norm": 0.8717501759529114, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 3480}, {"loss": 0.6188, "grad_norm": 0.6761979460716248, "learning_rate": 0.0002, "epoch": 3.8204707170224412, "step": 3490}, {"loss": 0.7601, "grad_norm": 1.0054380893707275, "learning_rate": 0.0002, "epoch": 3.8314176245210727, "step": 3500}, {"loss": 0.623, "grad_norm": 1.1224009990692139, "learning_rate": 0.0002, "epoch": 3.8423645320197046, "step": 3510}, {"loss": 0.6918, "grad_norm": 0.8997692465782166, "learning_rate": 0.0002, "epoch": 3.853311439518336, "step": 3520}, {"loss": 0.6357, "grad_norm": 1.0086902379989624, "learning_rate": 0.0002, "epoch": 3.8642583470169676, "step": 3530}, {"loss": 0.6379, "grad_norm": 0.772739589214325, "learning_rate": 0.0002, "epoch": 3.8752052545155995, "step": 3540}, {"loss": 0.7423, "grad_norm": 1.211774230003357, "learning_rate": 0.0002, "epoch": 3.886152162014231, "step": 3550}, {"loss": 0.7321, "grad_norm": 0.9572356939315796, "learning_rate": 0.0002, "epoch": 3.897099069512863, "step": 3560}, {"loss": 0.6836, "grad_norm": 0.7887842655181885, "learning_rate": 0.0002, "epoch": 3.9080459770114944, "step": 3570}, {"loss": 0.7576, "grad_norm": 0.7308389544487, "learning_rate": 0.0002, "epoch": 3.918992884510126, "step": 3580}, {"loss": 0.6001, "grad_norm": 1.0182650089263916, "learning_rate": 0.0002, "epoch": 3.9299397920087573, "step": 3590}, {"loss": 0.6942, "grad_norm": 0.8000147342681885, "learning_rate": 0.0002, "epoch": 3.9408866995073892, "step": 3600}, {"loss": 0.6244, "grad_norm": 0.7385728359222412, "learning_rate": 0.0002, "epoch": 3.9518336070060207, "step": 3610}, {"loss": 0.6718, "grad_norm": 0.9233261942863464, "learning_rate": 0.0002, "epoch": 3.9627805145046526, "step": 3620}, {"loss": 0.6508, "grad_norm": 0.8486751914024353, "learning_rate": 0.0002, "epoch": 3.973727422003284, "step": 3630}, {"loss": 0.6928, "grad_norm": 0.7593663334846497, "learning_rate": 0.0002, "epoch": 3.9846743295019156, "step": 3640}, {"loss": 0.6847, "grad_norm": 0.7885415554046631, "learning_rate": 0.0002, "epoch": 3.9956212370005475, "step": 3650}, {"eval_loss": 1.250312328338623, "eval_runtime": 46.0842, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.193, "epoch": 4.0, "step": 3654}, {"loss": 0.5547, "grad_norm": 0.6591703295707703, "learning_rate": 0.0002, "epoch": 4.006568144499179, "step": 3660}, {"loss": 0.5301, "grad_norm": 1.36927330493927, "learning_rate": 0.0002, "epoch": 4.017515051997811, "step": 3670}, {"loss": 0.4466, "grad_norm": 0.8106328845024109, "learning_rate": 0.0002, "epoch": 4.028461959496442, "step": 3680}, {"loss": 0.4861, "grad_norm": 0.7592712044715881, "learning_rate": 0.0002, "epoch": 4.039408866995074, "step": 3690}, {"loss": 0.5103, "grad_norm": 0.9518909454345703, "learning_rate": 0.0002, "epoch": 4.050355774493705, "step": 3700}, {"loss": 0.4638, "grad_norm": 0.7805967330932617, "learning_rate": 0.0002, "epoch": 4.061302681992337, "step": 3710}, {"loss": 0.4556, "grad_norm": 1.3146334886550903, "learning_rate": 0.0002, "epoch": 4.072249589490969, "step": 3720}, {"loss": 0.5635, "grad_norm": 1.1611138582229614, "learning_rate": 0.0002, "epoch": 4.083196496989601, "step": 3730}, {"loss": 0.3845, "grad_norm": 0.8173232078552246, "learning_rate": 0.0002, "epoch": 4.094143404488232, "step": 3740}, {"loss": 0.4911, "grad_norm": 0.7848323583602905, "learning_rate": 0.0002, "epoch": 4.105090311986864, "step": 3750}, {"loss": 0.4519, "grad_norm": 1.3183201551437378, "learning_rate": 0.0002, "epoch": 4.116037219485495, "step": 3760}, {"loss": 0.5083, "grad_norm": 1.1936529874801636, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 3770}, {"loss": 0.5208, "grad_norm": 1.1078993082046509, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 3780}, {"loss": 0.5928, "grad_norm": 1.107743263244629, "learning_rate": 0.0002, "epoch": 4.14887794198139, "step": 3790}, {"loss": 0.5112, "grad_norm": 0.7801875472068787, "learning_rate": 0.0002, "epoch": 4.159824849480022, "step": 3800}, {"loss": 0.4896, "grad_norm": 1.1328117847442627, "learning_rate": 0.0002, "epoch": 4.170771756978653, "step": 3810}, {"loss": 0.5645, "grad_norm": 1.4232193231582642, "learning_rate": 0.0002, "epoch": 4.181718664477285, "step": 3820}, {"loss": 0.5049, "grad_norm": 1.557416558265686, "learning_rate": 0.0002, "epoch": 4.192665571975917, "step": 3830}, {"loss": 0.4863, "grad_norm": 1.042923092842102, "learning_rate": 0.0002, "epoch": 4.203612479474549, "step": 3840}, {"loss": 0.3751, "grad_norm": 1.1801949739456177, "learning_rate": 0.0002, "epoch": 4.21455938697318, "step": 3850}, {"loss": 0.5063, "grad_norm": 0.9273753762245178, "learning_rate": 0.0002, "epoch": 4.225506294471812, "step": 3860}, {"loss": 0.5542, "grad_norm": 0.7681763768196106, "learning_rate": 0.0002, "epoch": 4.236453201970443, "step": 3870}, {"loss": 0.5971, "grad_norm": 0.9840841293334961, "learning_rate": 0.0002, "epoch": 4.2474001094690745, "step": 3880}, {"loss": 0.4648, "grad_norm": 1.0290725231170654, "learning_rate": 0.0002, "epoch": 4.258347016967707, "step": 3890}, {"loss": 0.4288, "grad_norm": 0.8059597611427307, "learning_rate": 0.0002, "epoch": 4.269293924466338, "step": 3900}, {"loss": 0.5103, "grad_norm": 0.9847467541694641, "learning_rate": 0.0002, "epoch": 4.28024083196497, "step": 3910}, {"loss": 0.4952, "grad_norm": 1.344044804573059, "learning_rate": 0.0002, "epoch": 4.291187739463601, "step": 3920}, {"loss": 0.4966, "grad_norm": 0.9174224138259888, "learning_rate": 0.0002, "epoch": 4.302134646962233, "step": 3930}, {"loss": 0.4944, "grad_norm": 1.1199711561203003, "learning_rate": 0.0002, "epoch": 4.313081554460865, "step": 3940}, {"loss": 0.4641, "grad_norm": 1.0120296478271484, "learning_rate": 0.0002, "epoch": 4.324028461959497, "step": 3950}, {"loss": 0.4723, "grad_norm": 1.091811180114746, "learning_rate": 0.0002, "epoch": 4.334975369458128, "step": 3960}, {"loss": 0.4627, "grad_norm": 1.0332133769989014, "learning_rate": 0.0002, "epoch": 4.34592227695676, "step": 3970}, {"loss": 0.4646, "grad_norm": 1.0785295963287354, "learning_rate": 0.0002, "epoch": 4.356869184455391, "step": 3980}, {"loss": 0.4909, "grad_norm": 1.0506969690322876, "learning_rate": 0.0002, "epoch": 4.3678160919540225, "step": 3990}, {"loss": 0.4776, "grad_norm": 1.047560691833496, "learning_rate": 0.0002, "epoch": 4.378762999452655, "step": 4000}, {"loss": 0.4549, "grad_norm": 0.9348800778388977, "learning_rate": 0.0002, "epoch": 4.389709906951286, "step": 4010}, {"loss": 0.5333, "grad_norm": 1.1563059091567993, "learning_rate": 0.0002, "epoch": 4.400656814449918, "step": 4020}, {"loss": 0.4952, "grad_norm": 1.001470923423767, "learning_rate": 0.0002, "epoch": 4.411603721948549, "step": 4030}, {"loss": 0.4972, "grad_norm": 1.309012532234192, "learning_rate": 0.0002, "epoch": 4.422550629447181, "step": 4040}, {"loss": 0.5078, "grad_norm": 0.7338925004005432, "learning_rate": 0.0002, "epoch": 4.433497536945813, "step": 4050}, {"loss": 0.4632, "grad_norm": 1.0398834943771362, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 4060}, {"loss": 0.6285, "grad_norm": 0.9728689193725586, "learning_rate": 0.0002, "epoch": 4.455391351943076, "step": 4070}, {"loss": 0.4778, "grad_norm": 1.247475028038025, "learning_rate": 0.0002, "epoch": 4.466338259441708, "step": 4080}, {"loss": 0.4813, "grad_norm": 1.1084578037261963, "learning_rate": 0.0002, "epoch": 4.477285166940339, "step": 4090}, {"loss": 0.5665, "grad_norm": 1.1619318723678589, "learning_rate": 0.0002, "epoch": 4.4882320744389705, "step": 4100}, {"loss": 0.5207, "grad_norm": 1.3456498384475708, "learning_rate": 0.0002, "epoch": 4.499178981937603, "step": 4110}, {"loss": 0.4876, "grad_norm": 0.9372991323471069, "learning_rate": 0.0002, "epoch": 4.510125889436234, "step": 4120}, {"loss": 0.5456, "grad_norm": 1.0071815252304077, "learning_rate": 0.0002, "epoch": 4.521072796934866, "step": 4130}, {"loss": 0.5589, "grad_norm": 1.190344214439392, "learning_rate": 0.0002, "epoch": 4.532019704433497, "step": 4140}, {"loss": 0.4852, "grad_norm": 0.9480887055397034, "learning_rate": 0.0002, "epoch": 4.542966611932129, "step": 4150}, {"loss": 0.5229, "grad_norm": 1.0252189636230469, "learning_rate": 0.0002, "epoch": 4.553913519430761, "step": 4160}, {"loss": 0.5253, "grad_norm": 0.7142013311386108, "learning_rate": 0.0002, "epoch": 4.564860426929393, "step": 4170}, {"loss": 0.4861, "grad_norm": 0.8937426805496216, "learning_rate": 0.0002, "epoch": 4.575807334428024, "step": 4180}, {"loss": 0.4773, "grad_norm": 0.8885005116462708, "learning_rate": 0.0002, "epoch": 4.586754241926656, "step": 4190}, {"loss": 0.4858, "grad_norm": 1.337663173675537, "learning_rate": 0.0002, "epoch": 4.597701149425287, "step": 4200}, {"loss": 0.5247, "grad_norm": 1.0475375652313232, "learning_rate": 0.0002, "epoch": 4.6086480569239185, "step": 4210}, {"loss": 0.5298, "grad_norm": 1.0081088542938232, "learning_rate": 0.0002, "epoch": 4.619594964422551, "step": 4220}, {"loss": 0.5042, "grad_norm": 0.7527595162391663, "learning_rate": 0.0002, "epoch": 4.630541871921182, "step": 4230}, {"loss": 0.5207, "grad_norm": 1.55559241771698, "learning_rate": 0.0002, "epoch": 4.641488779419814, "step": 4240}, {"loss": 0.5468, "grad_norm": 0.7967379689216614, "learning_rate": 0.0002, "epoch": 4.652435686918445, "step": 4250}, {"loss": 0.5328, "grad_norm": 0.898368775844574, "learning_rate": 0.0002, "epoch": 4.663382594417077, "step": 4260}, {"loss": 0.4706, "grad_norm": 1.1940776109695435, "learning_rate": 0.0002, "epoch": 4.674329501915709, "step": 4270}, {"loss": 0.5121, "grad_norm": 1.1817092895507812, "learning_rate": 0.0002, "epoch": 4.685276409414341, "step": 4280}, {"loss": 0.5758, "grad_norm": 0.9041520357131958, "learning_rate": 0.0002, "epoch": 4.696223316912972, "step": 4290}, {"loss": 0.5851, "grad_norm": 1.1280102729797363, "learning_rate": 0.0002, "epoch": 4.707170224411604, "step": 4300}, {"loss": 0.4891, "grad_norm": 1.357689619064331, "learning_rate": 0.0002, "epoch": 4.718117131910235, "step": 4310}, {"loss": 0.4704, "grad_norm": 1.056633472442627, "learning_rate": 0.0002, "epoch": 4.7290640394088665, "step": 4320}, {"loss": 0.5488, "grad_norm": 1.6520427465438843, "learning_rate": 0.0002, "epoch": 4.740010946907499, "step": 4330}, {"loss": 0.5131, "grad_norm": 1.153200626373291, "learning_rate": 0.0002, "epoch": 4.75095785440613, "step": 4340}, {"loss": 0.539, "grad_norm": 0.9346241354942322, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 4350}, {"loss": 0.4941, "grad_norm": 0.8628455996513367, "learning_rate": 0.0002, "epoch": 4.772851669403393, "step": 4360}, {"loss": 0.5167, "grad_norm": 1.3843916654586792, "learning_rate": 0.0002, "epoch": 4.783798576902025, "step": 4370}, {"loss": 0.4683, "grad_norm": 1.035574197769165, "learning_rate": 0.0002, "epoch": 4.794745484400657, "step": 4380}, {"loss": 0.5162, "grad_norm": 1.1868361234664917, "learning_rate": 0.0002, "epoch": 4.805692391899289, "step": 4390}, {"loss": 0.534, "grad_norm": 1.1307647228240967, "learning_rate": 0.0002, "epoch": 4.81663929939792, "step": 4400}, {"loss": 0.5567, "grad_norm": 0.9787724614143372, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 4410}, {"loss": 0.5185, "grad_norm": 1.0473824739456177, "learning_rate": 0.0002, "epoch": 4.838533114395183, "step": 4420}, {"loss": 0.6285, "grad_norm": 1.069069504737854, "learning_rate": 0.0002, "epoch": 4.8494800218938146, "step": 4430}, {"loss": 0.5267, "grad_norm": 1.4305680990219116, "learning_rate": 0.0002, "epoch": 4.860426929392447, "step": 4440}, {"loss": 0.5947, "grad_norm": 1.3679203987121582, "learning_rate": 0.0002, "epoch": 4.871373836891078, "step": 4450}, {"loss": 0.5135, "grad_norm": 0.8997844457626343, "learning_rate": 0.0002, "epoch": 4.88232074438971, "step": 4460}, {"loss": 0.5312, "grad_norm": 1.2758110761642456, "learning_rate": 0.0002, "epoch": 4.893267651888341, "step": 4470}, {"loss": 0.4914, "grad_norm": 0.8819465637207031, "learning_rate": 0.0002, "epoch": 4.904214559386973, "step": 4480}, {"loss": 0.5147, "grad_norm": 1.08329439163208, "learning_rate": 0.0002, "epoch": 4.915161466885605, "step": 4490}, {"loss": 0.5404, "grad_norm": 1.083461046218872, "learning_rate": 0.0002, "epoch": 4.926108374384237, "step": 4500}, {"loss": 0.5433, "grad_norm": 1.2387723922729492, "learning_rate": 0.0002, "epoch": 4.937055281882868, "step": 4510}, {"loss": 0.5624, "grad_norm": 0.8262293934822083, "learning_rate": 0.0002, "epoch": 4.9480021893815, "step": 4520}, {"loss": 0.504, "grad_norm": 1.2325191497802734, "learning_rate": 0.0002, "epoch": 4.958949096880131, "step": 4530}, {"loss": 0.5452, "grad_norm": 1.024614930152893, "learning_rate": 0.0002, "epoch": 4.9698960043787626, "step": 4540}, {"loss": 0.4752, "grad_norm": 1.3007521629333496, "learning_rate": 0.0002, "epoch": 4.980842911877395, "step": 4550}, {"loss": 0.4943, "grad_norm": 0.9823828339576721, "learning_rate": 0.0002, "epoch": 4.991789819376026, "step": 4560}]}
+{"epoch": 6.0, "step": 5481, "epoch_duration": 1359.439982175827, "total_accumulated_duration": 8170.820874452591, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}, {"eval_loss": 1.1722838878631592, "eval_runtime": 46.0829, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.194, "epoch": 2.9994526546250686, "step": 2740}, {"loss": 0.6443, "grad_norm": 0.6384502053260803, "learning_rate": 0.0002, "epoch": 3.0103995621237, "step": 2750}, {"loss": 0.7123, "grad_norm": 0.9928722381591797, "learning_rate": 0.0002, "epoch": 3.0213464696223316, "step": 2760}, {"loss": 0.6045, "grad_norm": 0.7813051342964172, "learning_rate": 0.0002, "epoch": 3.0322933771209635, "step": 2770}, {"loss": 0.6042, "grad_norm": 1.0202556848526, "learning_rate": 0.0002, "epoch": 3.043240284619595, "step": 2780}, {"loss": 0.6356, "grad_norm": 0.7581062316894531, "learning_rate": 0.0002, "epoch": 3.0541871921182264, "step": 2790}, {"loss": 0.6349, "grad_norm": 0.6252710223197937, "learning_rate": 0.0002, "epoch": 3.0651340996168583, "step": 2800}, {"loss": 0.645, "grad_norm": 0.7738662958145142, "learning_rate": 0.0002, "epoch": 3.07608100711549, "step": 2810}, {"loss": 0.627, "grad_norm": 0.7381885051727295, "learning_rate": 0.0002, "epoch": 3.0870279146141213, "step": 2820}, {"loss": 0.6371, "grad_norm": 0.9197564721107483, "learning_rate": 0.0002, "epoch": 3.097974822112753, "step": 2830}, {"loss": 0.723, "grad_norm": 1.000976800918579, "learning_rate": 0.0002, "epoch": 3.1089217296113847, "step": 2840}, {"loss": 0.6631, "grad_norm": 0.7559131383895874, "learning_rate": 0.0002, "epoch": 3.1198686371100166, "step": 2850}, {"loss": 0.6252, "grad_norm": 0.7213780879974365, "learning_rate": 0.0002, "epoch": 3.130815544608648, "step": 2860}, {"loss": 0.6501, "grad_norm": 0.945939838886261, "learning_rate": 0.0002, "epoch": 3.1417624521072796, "step": 2870}, {"loss": 0.6129, "grad_norm": 0.7277454137802124, "learning_rate": 0.0002, "epoch": 3.1527093596059115, "step": 2880}, {"loss": 0.6423, "grad_norm": 0.762026846408844, "learning_rate": 0.0002, "epoch": 3.163656267104543, "step": 2890}, {"loss": 0.5332, "grad_norm": 0.6471221446990967, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 2900}, {"loss": 0.7981, "grad_norm": 0.6018978357315063, "learning_rate": 0.0002, "epoch": 3.1855500821018063, "step": 2910}, {"loss": 0.7274, "grad_norm": 0.8607320785522461, "learning_rate": 0.0002, "epoch": 3.196496989600438, "step": 2920}, {"loss": 0.6139, "grad_norm": 0.8854126334190369, "learning_rate": 0.0002, "epoch": 3.2074438970990693, "step": 2930}, {"loss": 0.6485, "grad_norm": 0.6620870232582092, "learning_rate": 0.0002, "epoch": 3.218390804597701, "step": 2940}, {"loss": 0.6969, "grad_norm": 0.7377511858940125, "learning_rate": 0.0002, "epoch": 3.2293377120963327, "step": 2950}, {"loss": 0.6798, "grad_norm": 0.7803301811218262, "learning_rate": 0.0002, "epoch": 3.2402846195949646, "step": 2960}, {"loss": 0.6697, "grad_norm": 0.834061861038208, "learning_rate": 0.0002, "epoch": 3.251231527093596, "step": 2970}, {"loss": 0.6894, "grad_norm": 0.8496041893959045, "learning_rate": 0.0002, "epoch": 3.2621784345922276, "step": 2980}, {"loss": 0.6591, "grad_norm": 0.7967984676361084, "learning_rate": 0.0002, "epoch": 3.2731253420908595, "step": 2990}, {"loss": 0.7266, "grad_norm": 1.0207016468048096, "learning_rate": 0.0002, "epoch": 3.284072249589491, "step": 3000}, {"loss": 0.6586, "grad_norm": 0.7855771780014038, "learning_rate": 0.0002, "epoch": 3.2950191570881224, "step": 3010}, {"loss": 0.5711, "grad_norm": 0.9427546858787537, "learning_rate": 0.0002, "epoch": 3.3059660645867543, "step": 3020}, {"loss": 0.6277, "grad_norm": 0.823542594909668, "learning_rate": 0.0002, "epoch": 3.316912972085386, "step": 3030}, {"loss": 0.7109, "grad_norm": 0.9826635122299194, "learning_rate": 0.0002, "epoch": 3.3278598795840173, "step": 3040}, {"loss": 0.6564, "grad_norm": 0.7259827852249146, "learning_rate": 0.0002, "epoch": 3.338806787082649, "step": 3050}, {"loss": 0.653, "grad_norm": 0.7774739861488342, "learning_rate": 0.0002, "epoch": 3.3497536945812807, "step": 3060}, {"loss": 0.7529, "grad_norm": 0.7394293546676636, "learning_rate": 0.0002, "epoch": 3.3607006020799126, "step": 3070}, {"loss": 0.5987, "grad_norm": 0.9017578959465027, "learning_rate": 0.0002, "epoch": 3.371647509578544, "step": 3080}, {"loss": 0.6953, "grad_norm": 0.7451054453849792, "learning_rate": 0.0002, "epoch": 3.3825944170771756, "step": 3090}, {"loss": 0.6759, "grad_norm": 0.7321506142616272, "learning_rate": 0.0002, "epoch": 3.3935413245758075, "step": 3100}, {"loss": 0.6555, "grad_norm": 0.6721828579902649, "learning_rate": 0.0002, "epoch": 3.404488232074439, "step": 3110}, {"loss": 0.6559, "grad_norm": 0.774022102355957, "learning_rate": 0.0002, "epoch": 3.4154351395730704, "step": 3120}, {"loss": 0.7449, "grad_norm": 0.9143537282943726, "learning_rate": 0.0002, "epoch": 3.4263820470717024, "step": 3130}, {"loss": 0.6899, "grad_norm": 1.226087212562561, "learning_rate": 0.0002, "epoch": 3.437328954570334, "step": 3140}, {"loss": 0.6719, "grad_norm": 0.7545496225357056, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 3150}, {"loss": 0.6153, "grad_norm": 0.6515635848045349, "learning_rate": 0.0002, "epoch": 3.4592227695675972, "step": 3160}, {"loss": 0.6926, "grad_norm": 0.9297090172767639, "learning_rate": 0.0002, "epoch": 3.4701696770662287, "step": 3170}, {"loss": 0.6071, "grad_norm": 1.0130730867385864, "learning_rate": 0.0002, "epoch": 3.4811165845648606, "step": 3180}, {"loss": 0.5959, "grad_norm": 0.7654589414596558, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 3190}, {"loss": 0.7401, "grad_norm": 0.9954977631568909, "learning_rate": 0.0002, "epoch": 3.5030103995621236, "step": 3200}, {"loss": 0.6661, "grad_norm": 0.6027487516403198, "learning_rate": 0.0002, "epoch": 3.5139573070607555, "step": 3210}, {"loss": 0.6963, "grad_norm": 0.741770327091217, "learning_rate": 0.0002, "epoch": 3.524904214559387, "step": 3220}, {"loss": 0.8112, "grad_norm": 1.0534909963607788, "learning_rate": 0.0002, "epoch": 3.535851122058019, "step": 3230}, {"loss": 0.6813, "grad_norm": 0.937772274017334, "learning_rate": 0.0002, "epoch": 3.5467980295566504, "step": 3240}, {"loss": 0.6681, "grad_norm": 0.8504213690757751, "learning_rate": 0.0002, "epoch": 3.557744937055282, "step": 3250}, {"loss": 0.6436, "grad_norm": 0.7755007147789001, "learning_rate": 0.0002, "epoch": 3.5686918445539133, "step": 3260}, {"loss": 0.6213, "grad_norm": 1.0193358659744263, "learning_rate": 0.0002, "epoch": 3.5796387520525452, "step": 3270}, {"loss": 0.671, "grad_norm": 0.8440536856651306, "learning_rate": 0.0002, "epoch": 3.5905856595511767, "step": 3280}, {"loss": 0.6859, "grad_norm": 0.6195939183235168, "learning_rate": 0.0002, "epoch": 3.6015325670498086, "step": 3290}, {"loss": 0.7446, "grad_norm": 0.8608590960502625, "learning_rate": 0.0002, "epoch": 3.61247947454844, "step": 3300}, {"loss": 0.7301, "grad_norm": 0.6772327423095703, "learning_rate": 0.0002, "epoch": 3.6234263820470716, "step": 3310}, {"loss": 0.6298, "grad_norm": 0.8031839728355408, "learning_rate": 0.0002, "epoch": 3.6343732895457035, "step": 3320}, {"loss": 0.7041, "grad_norm": 0.6080502271652222, "learning_rate": 0.0002, "epoch": 3.645320197044335, "step": 3330}, {"loss": 0.7431, "grad_norm": 0.8007240891456604, "learning_rate": 0.0002, "epoch": 3.656267104542967, "step": 3340}, {"loss": 0.7446, "grad_norm": 0.8060704469680786, "learning_rate": 0.0002, "epoch": 3.6672140120415984, "step": 3350}, {"loss": 0.6304, "grad_norm": 0.7547586560249329, "learning_rate": 0.0002, "epoch": 3.67816091954023, "step": 3360}, {"loss": 0.7066, "grad_norm": 0.686851978302002, "learning_rate": 0.0002, "epoch": 3.6891078270388613, "step": 3370}, {"loss": 0.6748, "grad_norm": 0.9429075717926025, "learning_rate": 0.0002, "epoch": 3.7000547345374932, "step": 3380}, {"loss": 0.6673, "grad_norm": 0.7283591032028198, "learning_rate": 0.0002, "epoch": 3.7110016420361247, "step": 3390}, {"loss": 0.7502, "grad_norm": 0.8323085904121399, "learning_rate": 0.0002, "epoch": 3.7219485495347566, "step": 3400}, {"loss": 0.7779, "grad_norm": 0.8529590964317322, "learning_rate": 0.0002, "epoch": 3.732895457033388, "step": 3410}, {"loss": 0.6555, "grad_norm": 0.731752872467041, "learning_rate": 0.0002, "epoch": 3.7438423645320196, "step": 3420}, {"loss": 0.6928, "grad_norm": 0.8572278618812561, "learning_rate": 0.0002, "epoch": 3.7547892720306515, "step": 3430}, {"loss": 0.6215, "grad_norm": 0.7408691048622131, "learning_rate": 0.0002, "epoch": 3.765736179529283, "step": 3440}, {"loss": 0.622, "grad_norm": 0.7470445036888123, "learning_rate": 0.0002, "epoch": 3.776683087027915, "step": 3450}, {"loss": 0.7241, "grad_norm": 0.6806244254112244, "learning_rate": 0.0002, "epoch": 3.7876299945265464, "step": 3460}, {"loss": 0.7739, "grad_norm": 0.9129069447517395, "learning_rate": 0.0002, "epoch": 3.798576902025178, "step": 3470}, {"loss": 0.6826, "grad_norm": 0.8717501759529114, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 3480}, {"loss": 0.6188, "grad_norm": 0.6761979460716248, "learning_rate": 0.0002, "epoch": 3.8204707170224412, "step": 3490}, {"loss": 0.7601, "grad_norm": 1.0054380893707275, "learning_rate": 0.0002, "epoch": 3.8314176245210727, "step": 3500}, {"loss": 0.623, "grad_norm": 1.1224009990692139, "learning_rate": 0.0002, "epoch": 3.8423645320197046, "step": 3510}, {"loss": 0.6918, "grad_norm": 0.8997692465782166, "learning_rate": 0.0002, "epoch": 3.853311439518336, "step": 3520}, {"loss": 0.6357, "grad_norm": 1.0086902379989624, "learning_rate": 0.0002, "epoch": 3.8642583470169676, "step": 3530}, {"loss": 0.6379, "grad_norm": 0.772739589214325, "learning_rate": 0.0002, "epoch": 3.8752052545155995, "step": 3540}, {"loss": 0.7423, "grad_norm": 1.211774230003357, "learning_rate": 0.0002, "epoch": 3.886152162014231, "step": 3550}, {"loss": 0.7321, "grad_norm": 0.9572356939315796, "learning_rate": 0.0002, "epoch": 3.897099069512863, "step": 3560}, {"loss": 0.6836, "grad_norm": 0.7887842655181885, "learning_rate": 0.0002, "epoch": 3.9080459770114944, "step": 3570}, {"loss": 0.7576, "grad_norm": 0.7308389544487, "learning_rate": 0.0002, "epoch": 3.918992884510126, "step": 3580}, {"loss": 0.6001, "grad_norm": 1.0182650089263916, "learning_rate": 0.0002, "epoch": 3.9299397920087573, "step": 3590}, {"loss": 0.6942, "grad_norm": 0.8000147342681885, "learning_rate": 0.0002, "epoch": 3.9408866995073892, "step": 3600}, {"loss": 0.6244, "grad_norm": 0.7385728359222412, "learning_rate": 0.0002, "epoch": 3.9518336070060207, "step": 3610}, {"loss": 0.6718, "grad_norm": 0.9233261942863464, "learning_rate": 0.0002, "epoch": 3.9627805145046526, "step": 3620}, {"loss": 0.6508, "grad_norm": 0.8486751914024353, "learning_rate": 0.0002, "epoch": 3.973727422003284, "step": 3630}, {"loss": 0.6928, "grad_norm": 0.7593663334846497, "learning_rate": 0.0002, "epoch": 3.9846743295019156, "step": 3640}, {"loss": 0.6847, "grad_norm": 0.7885415554046631, "learning_rate": 0.0002, "epoch": 3.9956212370005475, "step": 3650}, {"eval_loss": 1.250312328338623, "eval_runtime": 46.0842, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.193, "epoch": 4.0, "step": 3654}, {"loss": 0.5547, "grad_norm": 0.6591703295707703, "learning_rate": 0.0002, "epoch": 4.006568144499179, "step": 3660}, {"loss": 0.5301, "grad_norm": 1.36927330493927, "learning_rate": 0.0002, "epoch": 4.017515051997811, "step": 3670}, {"loss": 0.4466, "grad_norm": 0.8106328845024109, "learning_rate": 0.0002, "epoch": 4.028461959496442, "step": 3680}, {"loss": 0.4861, "grad_norm": 0.7592712044715881, "learning_rate": 0.0002, "epoch": 4.039408866995074, "step": 3690}, {"loss": 0.5103, "grad_norm": 0.9518909454345703, "learning_rate": 0.0002, "epoch": 4.050355774493705, "step": 3700}, {"loss": 0.4638, "grad_norm": 0.7805967330932617, "learning_rate": 0.0002, "epoch": 4.061302681992337, "step": 3710}, {"loss": 0.4556, "grad_norm": 1.3146334886550903, "learning_rate": 0.0002, "epoch": 4.072249589490969, "step": 3720}, {"loss": 0.5635, "grad_norm": 1.1611138582229614, "learning_rate": 0.0002, "epoch": 4.083196496989601, "step": 3730}, {"loss": 0.3845, "grad_norm": 0.8173232078552246, "learning_rate": 0.0002, "epoch": 4.094143404488232, "step": 3740}, {"loss": 0.4911, "grad_norm": 0.7848323583602905, "learning_rate": 0.0002, "epoch": 4.105090311986864, "step": 3750}, {"loss": 0.4519, "grad_norm": 1.3183201551437378, "learning_rate": 0.0002, "epoch": 4.116037219485495, "step": 3760}, {"loss": 0.5083, "grad_norm": 1.1936529874801636, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 3770}, {"loss": 0.5208, "grad_norm": 1.1078993082046509, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 3780}, {"loss": 0.5928, "grad_norm": 1.107743263244629, "learning_rate": 0.0002, "epoch": 4.14887794198139, "step": 3790}, {"loss": 0.5112, "grad_norm": 0.7801875472068787, "learning_rate": 0.0002, "epoch": 4.159824849480022, "step": 3800}, {"loss": 0.4896, "grad_norm": 1.1328117847442627, "learning_rate": 0.0002, "epoch": 4.170771756978653, "step": 3810}, {"loss": 0.5645, "grad_norm": 1.4232193231582642, "learning_rate": 0.0002, "epoch": 4.181718664477285, "step": 3820}, {"loss": 0.5049, "grad_norm": 1.557416558265686, "learning_rate": 0.0002, "epoch": 4.192665571975917, "step": 3830}, {"loss": 0.4863, "grad_norm": 1.042923092842102, "learning_rate": 0.0002, "epoch": 4.203612479474549, "step": 3840}, {"loss": 0.3751, "grad_norm": 1.1801949739456177, "learning_rate": 0.0002, "epoch": 4.21455938697318, "step": 3850}, {"loss": 0.5063, "grad_norm": 0.9273753762245178, "learning_rate": 0.0002, "epoch": 4.225506294471812, "step": 3860}, {"loss": 0.5542, "grad_norm": 0.7681763768196106, "learning_rate": 0.0002, "epoch": 4.236453201970443, "step": 3870}, {"loss": 0.5971, "grad_norm": 0.9840841293334961, "learning_rate": 0.0002, "epoch": 4.2474001094690745, "step": 3880}, {"loss": 0.4648, "grad_norm": 1.0290725231170654, "learning_rate": 0.0002, "epoch": 4.258347016967707, "step": 3890}, {"loss": 0.4288, "grad_norm": 0.8059597611427307, "learning_rate": 0.0002, "epoch": 4.269293924466338, "step": 3900}, {"loss": 0.5103, "grad_norm": 0.9847467541694641, "learning_rate": 0.0002, "epoch": 4.28024083196497, "step": 3910}, {"loss": 0.4952, "grad_norm": 1.344044804573059, "learning_rate": 0.0002, "epoch": 4.291187739463601, "step": 3920}, {"loss": 0.4966, "grad_norm": 0.9174224138259888, "learning_rate": 0.0002, "epoch": 4.302134646962233, "step": 3930}, {"loss": 0.4944, "grad_norm": 1.1199711561203003, "learning_rate": 0.0002, "epoch": 4.313081554460865, "step": 3940}, {"loss": 0.4641, "grad_norm": 1.0120296478271484, "learning_rate": 0.0002, "epoch": 4.324028461959497, "step": 3950}, {"loss": 0.4723, "grad_norm": 1.091811180114746, "learning_rate": 0.0002, "epoch": 4.334975369458128, "step": 3960}, {"loss": 0.4627, "grad_norm": 1.0332133769989014, "learning_rate": 0.0002, "epoch": 4.34592227695676, "step": 3970}, {"loss": 0.4646, "grad_norm": 1.0785295963287354, "learning_rate": 0.0002, "epoch": 4.356869184455391, "step": 3980}, {"loss": 0.4909, "grad_norm": 1.0506969690322876, "learning_rate": 0.0002, "epoch": 4.3678160919540225, "step": 3990}, {"loss": 0.4776, "grad_norm": 1.047560691833496, "learning_rate": 0.0002, "epoch": 4.378762999452655, "step": 4000}, {"loss": 0.4549, "grad_norm": 0.9348800778388977, "learning_rate": 0.0002, "epoch": 4.389709906951286, "step": 4010}, {"loss": 0.5333, "grad_norm": 1.1563059091567993, "learning_rate": 0.0002, "epoch": 4.400656814449918, "step": 4020}, {"loss": 0.4952, "grad_norm": 1.001470923423767, "learning_rate": 0.0002, "epoch": 4.411603721948549, "step": 4030}, {"loss": 0.4972, "grad_norm": 1.309012532234192, "learning_rate": 0.0002, "epoch": 4.422550629447181, "step": 4040}, {"loss": 0.5078, "grad_norm": 0.7338925004005432, "learning_rate": 0.0002, "epoch": 4.433497536945813, "step": 4050}, {"loss": 0.4632, "grad_norm": 1.0398834943771362, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 4060}, {"loss": 0.6285, "grad_norm": 0.9728689193725586, "learning_rate": 0.0002, "epoch": 4.455391351943076, "step": 4070}, {"loss": 0.4778, "grad_norm": 1.247475028038025, "learning_rate": 0.0002, "epoch": 4.466338259441708, "step": 4080}, {"loss": 0.4813, "grad_norm": 1.1084578037261963, "learning_rate": 0.0002, "epoch": 4.477285166940339, "step": 4090}, {"loss": 0.5665, "grad_norm": 1.1619318723678589, "learning_rate": 0.0002, "epoch": 4.4882320744389705, "step": 4100}, {"loss": 0.5207, "grad_norm": 1.3456498384475708, "learning_rate": 0.0002, "epoch": 4.499178981937603, "step": 4110}, {"loss": 0.4876, "grad_norm": 0.9372991323471069, "learning_rate": 0.0002, "epoch": 4.510125889436234, "step": 4120}, {"loss": 0.5456, "grad_norm": 1.0071815252304077, "learning_rate": 0.0002, "epoch": 4.521072796934866, "step": 4130}, {"loss": 0.5589, "grad_norm": 1.190344214439392, "learning_rate": 0.0002, "epoch": 4.532019704433497, "step": 4140}, {"loss": 0.4852, "grad_norm": 0.9480887055397034, "learning_rate": 0.0002, "epoch": 4.542966611932129, "step": 4150}, {"loss": 0.5229, "grad_norm": 1.0252189636230469, "learning_rate": 0.0002, "epoch": 4.553913519430761, "step": 4160}, {"loss": 0.5253, "grad_norm": 0.7142013311386108, "learning_rate": 0.0002, "epoch": 4.564860426929393, "step": 4170}, {"loss": 0.4861, "grad_norm": 0.8937426805496216, "learning_rate": 0.0002, "epoch": 4.575807334428024, "step": 4180}, {"loss": 0.4773, "grad_norm": 0.8885005116462708, "learning_rate": 0.0002, "epoch": 4.586754241926656, "step": 4190}, {"loss": 0.4858, "grad_norm": 1.337663173675537, "learning_rate": 0.0002, "epoch": 4.597701149425287, "step": 4200}, {"loss": 0.5247, "grad_norm": 1.0475375652313232, "learning_rate": 0.0002, "epoch": 4.6086480569239185, "step": 4210}, {"loss": 0.5298, "grad_norm": 1.0081088542938232, "learning_rate": 0.0002, "epoch": 4.619594964422551, "step": 4220}, {"loss": 0.5042, "grad_norm": 0.7527595162391663, "learning_rate": 0.0002, "epoch": 4.630541871921182, "step": 4230}, {"loss": 0.5207, "grad_norm": 1.55559241771698, "learning_rate": 0.0002, "epoch": 4.641488779419814, "step": 4240}, {"loss": 0.5468, "grad_norm": 0.7967379689216614, "learning_rate": 0.0002, "epoch": 4.652435686918445, "step": 4250}, {"loss": 0.5328, "grad_norm": 0.898368775844574, "learning_rate": 0.0002, "epoch": 4.663382594417077, "step": 4260}, {"loss": 0.4706, "grad_norm": 1.1940776109695435, "learning_rate": 0.0002, "epoch": 4.674329501915709, "step": 4270}, {"loss": 0.5121, "grad_norm": 1.1817092895507812, "learning_rate": 0.0002, "epoch": 4.685276409414341, "step": 4280}, {"loss": 0.5758, "grad_norm": 0.9041520357131958, "learning_rate": 0.0002, "epoch": 4.696223316912972, "step": 4290}, {"loss": 0.5851, "grad_norm": 1.1280102729797363, "learning_rate": 0.0002, "epoch": 4.707170224411604, "step": 4300}, {"loss": 0.4891, "grad_norm": 1.357689619064331, "learning_rate": 0.0002, "epoch": 4.718117131910235, "step": 4310}, {"loss": 0.4704, "grad_norm": 1.056633472442627, "learning_rate": 0.0002, "epoch": 4.7290640394088665, "step": 4320}, {"loss": 0.5488, "grad_norm": 1.6520427465438843, "learning_rate": 0.0002, "epoch": 4.740010946907499, "step": 4330}, {"loss": 0.5131, "grad_norm": 1.153200626373291, "learning_rate": 0.0002, "epoch": 4.75095785440613, "step": 4340}, {"loss": 0.539, "grad_norm": 0.9346241354942322, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 4350}, {"loss": 0.4941, "grad_norm": 0.8628455996513367, "learning_rate": 0.0002, "epoch": 4.772851669403393, "step": 4360}, {"loss": 0.5167, "grad_norm": 1.3843916654586792, "learning_rate": 0.0002, "epoch": 4.783798576902025, "step": 4370}, {"loss": 0.4683, "grad_norm": 1.035574197769165, "learning_rate": 0.0002, "epoch": 4.794745484400657, "step": 4380}, {"loss": 0.5162, "grad_norm": 1.1868361234664917, "learning_rate": 0.0002, "epoch": 4.805692391899289, "step": 4390}, {"loss": 0.534, "grad_norm": 1.1307647228240967, "learning_rate": 0.0002, "epoch": 4.81663929939792, "step": 4400}, {"loss": 0.5567, "grad_norm": 0.9787724614143372, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 4410}, {"loss": 0.5185, "grad_norm": 1.0473824739456177, "learning_rate": 0.0002, "epoch": 4.838533114395183, "step": 4420}, {"loss": 0.6285, "grad_norm": 1.069069504737854, "learning_rate": 0.0002, "epoch": 4.8494800218938146, "step": 4430}, {"loss": 0.5267, "grad_norm": 1.4305680990219116, "learning_rate": 0.0002, "epoch": 4.860426929392447, "step": 4440}, {"loss": 0.5947, "grad_norm": 1.3679203987121582, "learning_rate": 0.0002, "epoch": 4.871373836891078, "step": 4450}, {"loss": 0.5135, "grad_norm": 0.8997844457626343, "learning_rate": 0.0002, "epoch": 4.88232074438971, "step": 4460}, {"loss": 0.5312, "grad_norm": 1.2758110761642456, "learning_rate": 0.0002, "epoch": 4.893267651888341, "step": 4470}, {"loss": 0.4914, "grad_norm": 0.8819465637207031, "learning_rate": 0.0002, "epoch": 4.904214559386973, "step": 4480}, {"loss": 0.5147, "grad_norm": 1.08329439163208, "learning_rate": 0.0002, "epoch": 4.915161466885605, "step": 4490}, {"loss": 0.5404, "grad_norm": 1.083461046218872, "learning_rate": 0.0002, "epoch": 4.926108374384237, "step": 4500}, {"loss": 0.5433, "grad_norm": 1.2387723922729492, "learning_rate": 0.0002, "epoch": 4.937055281882868, "step": 4510}, {"loss": 0.5624, "grad_norm": 0.8262293934822083, "learning_rate": 0.0002, "epoch": 4.9480021893815, "step": 4520}, {"loss": 0.504, "grad_norm": 1.2325191497802734, "learning_rate": 0.0002, "epoch": 4.958949096880131, "step": 4530}, {"loss": 0.5452, "grad_norm": 1.024614930152893, "learning_rate": 0.0002, "epoch": 4.9698960043787626, "step": 4540}, {"loss": 0.4752, "grad_norm": 1.3007521629333496, "learning_rate": 0.0002, "epoch": 4.980842911877395, "step": 4550}, {"loss": 0.4943, "grad_norm": 0.9823828339576721, "learning_rate": 0.0002, "epoch": 4.991789819376026, "step": 4560}, {"eval_loss": 1.3920727968215942, "eval_runtime": 46.0764, "eval_samples_per_second": 9.463, "eval_steps_per_second": 1.194, "epoch": 4.999452654625069, "step": 4567}, {"loss": 0.545, "grad_norm": 1.1478906869888306, "learning_rate": 0.0002, "epoch": 5.002736726874658, "step": 4570}, {"loss": 0.372, "grad_norm": 1.0533705949783325, "learning_rate": 0.0002, "epoch": 5.013683634373289, "step": 4580}, {"loss": 0.3313, "grad_norm": 1.268900752067566, "learning_rate": 0.0002, "epoch": 5.024630541871921, "step": 4590}, {"loss": 0.3482, "grad_norm": 1.222652554512024, "learning_rate": 0.0002, "epoch": 5.035577449370553, "step": 4600}, {"loss": 0.3195, "grad_norm": 1.5093127489089966, "learning_rate": 0.0002, "epoch": 5.046524356869185, "step": 4610}, {"loss": 0.3569, "grad_norm": 1.2372499704360962, "learning_rate": 0.0002, "epoch": 5.057471264367816, "step": 4620}, {"loss": 0.3206, "grad_norm": 0.8422666192054749, "learning_rate": 0.0002, "epoch": 5.068418171866448, "step": 4630}, {"loss": 0.3115, "grad_norm": 1.1451770067214966, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 4640}, {"loss": 0.3305, "grad_norm": 1.2074557542800903, "learning_rate": 0.0002, "epoch": 5.090311986863711, "step": 4650}, {"loss": 0.3012, "grad_norm": 1.429150104522705, "learning_rate": 0.0002, "epoch": 5.101258894362343, "step": 4660}, {"loss": 0.3229, "grad_norm": 1.0353610515594482, "learning_rate": 0.0002, "epoch": 5.112205801860974, "step": 4670}, {"loss": 0.402, "grad_norm": 1.2845979928970337, "learning_rate": 0.0002, "epoch": 5.123152709359606, "step": 4680}, {"loss": 0.383, "grad_norm": 1.3790186643600464, "learning_rate": 0.0002, "epoch": 5.134099616858237, "step": 4690}, {"loss": 0.2951, "grad_norm": 1.3182239532470703, "learning_rate": 0.0002, "epoch": 5.145046524356869, "step": 4700}, {"loss": 0.4074, "grad_norm": 1.5249626636505127, "learning_rate": 0.0002, "epoch": 5.155993431855501, "step": 4710}, {"loss": 0.3703, "grad_norm": 1.2492733001708984, "learning_rate": 0.0002, "epoch": 5.166940339354133, "step": 4720}, {"loss": 0.3411, "grad_norm": 1.4455480575561523, "learning_rate": 0.0002, "epoch": 5.177887246852764, "step": 4730}, {"loss": 0.3996, "grad_norm": 1.2191482782363892, "learning_rate": 0.0002, "epoch": 5.188834154351396, "step": 4740}, {"loss": 0.3785, "grad_norm": 1.4707951545715332, "learning_rate": 0.0002, "epoch": 5.199781061850027, "step": 4750}, {"loss": 0.3516, "grad_norm": 1.3473678827285767, "learning_rate": 0.0002, "epoch": 5.210727969348659, "step": 4760}, {"loss": 0.3266, "grad_norm": 1.0479670763015747, "learning_rate": 0.0002, "epoch": 5.221674876847291, "step": 4770}, {"loss": 0.3976, "grad_norm": 1.299096703529358, "learning_rate": 0.0002, "epoch": 5.232621784345922, "step": 4780}, {"loss": 0.3266, "grad_norm": 1.2820168733596802, "learning_rate": 0.0002, "epoch": 5.243568691844554, "step": 4790}, {"loss": 0.3347, "grad_norm": 1.3818004131317139, "learning_rate": 0.0002, "epoch": 5.254515599343185, "step": 4800}, {"loss": 0.3761, "grad_norm": 1.2898736000061035, "learning_rate": 0.0002, "epoch": 5.265462506841817, "step": 4810}, {"loss": 0.3694, "grad_norm": 1.1761468648910522, "learning_rate": 0.0002, "epoch": 5.276409414340449, "step": 4820}, {"loss": 0.3806, "grad_norm": 1.7155952453613281, "learning_rate": 0.0002, "epoch": 5.287356321839081, "step": 4830}, {"loss": 0.322, "grad_norm": 0.9103642106056213, "learning_rate": 0.0002, "epoch": 5.298303229337712, "step": 4840}, {"loss": 0.3516, "grad_norm": 1.013015627861023, "learning_rate": 0.0002, "epoch": 5.309250136836344, "step": 4850}, {"loss": 0.4297, "grad_norm": 1.390471339225769, "learning_rate": 0.0002, "epoch": 5.320197044334975, "step": 4860}, {"loss": 0.4098, "grad_norm": 1.129770278930664, "learning_rate": 0.0002, "epoch": 5.331143951833607, "step": 4870}, {"loss": 0.4227, "grad_norm": 1.1461067199707031, "learning_rate": 0.0002, "epoch": 5.342090859332239, "step": 4880}, {"loss": 0.288, "grad_norm": 1.3587424755096436, "learning_rate": 0.0002, "epoch": 5.35303776683087, "step": 4890}, {"loss": 0.3604, "grad_norm": 1.6897879838943481, "learning_rate": 0.0002, "epoch": 5.363984674329502, "step": 4900}, {"loss": 0.3887, "grad_norm": 0.9298055768013, "learning_rate": 0.0002, "epoch": 5.374931581828133, "step": 4910}, {"loss": 0.3371, "grad_norm": 1.0006917715072632, "learning_rate": 0.0002, "epoch": 5.385878489326765, "step": 4920}, {"loss": 0.3992, "grad_norm": 1.232581377029419, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 4930}, {"loss": 0.3456, "grad_norm": 1.0822620391845703, "learning_rate": 0.0002, "epoch": 5.407772304324029, "step": 4940}, {"loss": 0.3806, "grad_norm": 1.3648720979690552, "learning_rate": 0.0002, "epoch": 5.41871921182266, "step": 4950}, {"loss": 0.3959, "grad_norm": 1.3220354318618774, "learning_rate": 0.0002, "epoch": 5.429666119321292, "step": 4960}, {"loss": 0.3278, "grad_norm": 1.1106271743774414, "learning_rate": 0.0002, "epoch": 5.440613026819923, "step": 4970}, {"loss": 0.3812, "grad_norm": 1.6058908700942993, "learning_rate": 0.0002, "epoch": 5.451559934318555, "step": 4980}, {"loss": 0.3905, "grad_norm": 1.1065930128097534, "learning_rate": 0.0002, "epoch": 5.462506841817187, "step": 4990}, {"loss": 0.4058, "grad_norm": 1.3896466493606567, "learning_rate": 0.0002, "epoch": 5.473453749315818, "step": 5000}, {"loss": 0.4122, "grad_norm": 1.0437148809432983, "learning_rate": 0.0002, "epoch": 5.48440065681445, "step": 5010}, {"loss": 0.4065, "grad_norm": 1.2347718477249146, "learning_rate": 0.0002, "epoch": 5.495347564313081, "step": 5020}, {"loss": 0.3586, "grad_norm": 1.1174284219741821, "learning_rate": 0.0002, "epoch": 5.506294471811713, "step": 5030}, {"loss": 0.3576, "grad_norm": 1.2580941915512085, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 5040}, {"loss": 0.3809, "grad_norm": 1.451090931892395, "learning_rate": 0.0002, "epoch": 5.528188286808977, "step": 5050}, {"loss": 0.3645, "grad_norm": 1.4688365459442139, "learning_rate": 0.0002, "epoch": 5.539135194307608, "step": 5060}, {"loss": 0.4431, "grad_norm": 1.1625734567642212, "learning_rate": 0.0002, "epoch": 5.55008210180624, "step": 5070}, {"loss": 0.3972, "grad_norm": 0.9332265257835388, "learning_rate": 0.0002, "epoch": 5.561029009304871, "step": 5080}, {"loss": 0.4, "grad_norm": 1.5635273456573486, "learning_rate": 0.0002, "epoch": 5.571975916803503, "step": 5090}, {"loss": 0.3651, "grad_norm": 1.3420509099960327, "learning_rate": 0.0002, "epoch": 5.582922824302135, "step": 5100}, {"loss": 0.3717, "grad_norm": 1.5826557874679565, "learning_rate": 0.0002, "epoch": 5.593869731800766, "step": 5110}, {"loss": 0.4256, "grad_norm": 1.5737065076828003, "learning_rate": 0.0002, "epoch": 5.604816639299398, "step": 5120}, {"loss": 0.39, "grad_norm": 1.3812499046325684, "learning_rate": 0.0002, "epoch": 5.615763546798029, "step": 5130}, {"loss": 0.3891, "grad_norm": 1.362833023071289, "learning_rate": 0.0002, "epoch": 5.626710454296661, "step": 5140}, {"loss": 0.455, "grad_norm": 1.7667874097824097, "learning_rate": 0.0002, "epoch": 5.637657361795293, "step": 5150}, {"loss": 0.4264, "grad_norm": 1.2661789655685425, "learning_rate": 0.0002, "epoch": 5.648604269293925, "step": 5160}, {"loss": 0.3261, "grad_norm": 1.2076870203018188, "learning_rate": 0.0002, "epoch": 5.659551176792556, "step": 5170}, {"loss": 0.372, "grad_norm": 1.2431524991989136, "learning_rate": 0.0002, "epoch": 5.670498084291188, "step": 5180}, {"loss": 0.4092, "grad_norm": 1.2216639518737793, "learning_rate": 0.0002, "epoch": 5.681444991789819, "step": 5190}, {"loss": 0.4171, "grad_norm": 0.9259352684020996, "learning_rate": 0.0002, "epoch": 5.692391899288451, "step": 5200}, {"loss": 0.3875, "grad_norm": 1.7929338216781616, "learning_rate": 0.0002, "epoch": 5.703338806787083, "step": 5210}, {"loss": 0.4424, "grad_norm": 1.4048460721969604, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 5220}, {"loss": 0.3758, "grad_norm": 1.306874394416809, "learning_rate": 0.0002, "epoch": 5.725232621784346, "step": 5230}, {"loss": 0.3889, "grad_norm": 1.3137940168380737, "learning_rate": 0.0002, "epoch": 5.736179529282977, "step": 5240}, {"loss": 0.4804, "grad_norm": 1.1376476287841797, "learning_rate": 0.0002, "epoch": 5.747126436781609, "step": 5250}, {"loss": 0.377, "grad_norm": 1.450939416885376, "learning_rate": 0.0002, "epoch": 5.758073344280241, "step": 5260}, {"loss": 0.4732, "grad_norm": 0.983195960521698, "learning_rate": 0.0002, "epoch": 5.769020251778873, "step": 5270}, {"loss": 0.4041, "grad_norm": 1.66558837890625, "learning_rate": 0.0002, "epoch": 5.779967159277504, "step": 5280}, {"loss": 0.3643, "grad_norm": 0.9789204597473145, "learning_rate": 0.0002, "epoch": 5.790914066776136, "step": 5290}, {"loss": 0.3776, "grad_norm": 1.2110556364059448, "learning_rate": 0.0002, "epoch": 5.801860974274767, "step": 5300}, {"loss": 0.4049, "grad_norm": 1.3799304962158203, "learning_rate": 0.0002, "epoch": 5.812807881773399, "step": 5310}, {"loss": 0.4362, "grad_norm": 1.0570626258850098, "learning_rate": 0.0002, "epoch": 5.823754789272031, "step": 5320}, {"loss": 0.4716, "grad_norm": 1.4654436111450195, "learning_rate": 0.0002, "epoch": 5.834701696770662, "step": 5330}, {"loss": 0.4048, "grad_norm": 1.5216940641403198, "learning_rate": 0.0002, "epoch": 5.845648604269294, "step": 5340}, {"loss": 0.3848, "grad_norm": 1.018646001815796, "learning_rate": 0.0002, "epoch": 5.856595511767925, "step": 5350}, {"loss": 0.3705, "grad_norm": 1.028951644897461, "learning_rate": 0.0002, "epoch": 5.867542419266557, "step": 5360}, {"loss": 0.4213, "grad_norm": 2.571263313293457, "learning_rate": 0.0002, "epoch": 5.878489326765189, "step": 5370}, {"loss": 0.3647, "grad_norm": 1.3323984146118164, "learning_rate": 0.0002, "epoch": 5.889436234263821, "step": 5380}, {"loss": 0.4085, "grad_norm": 1.4317777156829834, "learning_rate": 0.0002, "epoch": 5.900383141762452, "step": 5390}, {"loss": 0.4254, "grad_norm": 1.4289140701293945, "learning_rate": 0.0002, "epoch": 5.911330049261084, "step": 5400}, {"loss": 0.3993, "grad_norm": 1.3130780458450317, "learning_rate": 0.0002, "epoch": 5.922276956759715, "step": 5410}, {"loss": 0.4025, "grad_norm": 1.3979902267456055, "learning_rate": 0.0002, "epoch": 5.933223864258347, "step": 5420}, {"loss": 0.3997, "grad_norm": 1.1827352046966553, "learning_rate": 0.0002, "epoch": 5.944170771756979, "step": 5430}, {"loss": 0.4163, "grad_norm": 1.1672080755233765, "learning_rate": 0.0002, "epoch": 5.95511767925561, "step": 5440}, {"loss": 0.4425, "grad_norm": 1.0949620008468628, "learning_rate": 0.0002, "epoch": 5.966064586754242, "step": 5450}, {"loss": 0.4219, "grad_norm": 1.3183925151824951, "learning_rate": 0.0002, "epoch": 5.977011494252873, "step": 5460}, {"loss": 0.4171, "grad_norm": 1.096198320388794, "learning_rate": 0.0002, "epoch": 5.987958401751505, "step": 5470}, {"loss": 0.3886, "grad_norm": 1.2601423263549805, "learning_rate": 0.0002, "epoch": 5.998905309250137, "step": 5480}]}
+{"epoch": 6.999452654625069, "step": 6394, "epoch_duration": 1359.7705018520355, "total_accumulated_duration": 9530.591376304626, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}, {"eval_loss": 1.1722838878631592, "eval_runtime": 46.0829, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.194, "epoch": 2.9994526546250686, "step": 2740}, {"loss": 0.6443, "grad_norm": 0.6384502053260803, "learning_rate": 0.0002, "epoch": 3.0103995621237, "step": 2750}, {"loss": 0.7123, "grad_norm": 0.9928722381591797, "learning_rate": 0.0002, "epoch": 3.0213464696223316, "step": 2760}, {"loss": 0.6045, "grad_norm": 0.7813051342964172, "learning_rate": 0.0002, "epoch": 3.0322933771209635, "step": 2770}, {"loss": 0.6042, "grad_norm": 1.0202556848526, "learning_rate": 0.0002, "epoch": 3.043240284619595, "step": 2780}, {"loss": 0.6356, "grad_norm": 0.7581062316894531, "learning_rate": 0.0002, "epoch": 3.0541871921182264, "step": 2790}, {"loss": 0.6349, "grad_norm": 0.6252710223197937, "learning_rate": 0.0002, "epoch": 3.0651340996168583, "step": 2800}, {"loss": 0.645, "grad_norm": 0.7738662958145142, "learning_rate": 0.0002, "epoch": 3.07608100711549, "step": 2810}, {"loss": 0.627, "grad_norm": 0.7381885051727295, "learning_rate": 0.0002, "epoch": 3.0870279146141213, "step": 2820}, {"loss": 0.6371, "grad_norm": 0.9197564721107483, "learning_rate": 0.0002, "epoch": 3.097974822112753, "step": 2830}, {"loss": 0.723, "grad_norm": 1.000976800918579, "learning_rate": 0.0002, "epoch": 3.1089217296113847, "step": 2840}, {"loss": 0.6631, "grad_norm": 0.7559131383895874, "learning_rate": 0.0002, "epoch": 3.1198686371100166, "step": 2850}, {"loss": 0.6252, "grad_norm": 0.7213780879974365, "learning_rate": 0.0002, "epoch": 3.130815544608648, "step": 2860}, {"loss": 0.6501, "grad_norm": 0.945939838886261, "learning_rate": 0.0002, "epoch": 3.1417624521072796, "step": 2870}, {"loss": 0.6129, "grad_norm": 0.7277454137802124, "learning_rate": 0.0002, "epoch": 3.1527093596059115, "step": 2880}, {"loss": 0.6423, "grad_norm": 0.762026846408844, "learning_rate": 0.0002, "epoch": 3.163656267104543, "step": 2890}, {"loss": 0.5332, "grad_norm": 0.6471221446990967, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 2900}, {"loss": 0.7981, "grad_norm": 0.6018978357315063, "learning_rate": 0.0002, "epoch": 3.1855500821018063, "step": 2910}, {"loss": 0.7274, "grad_norm": 0.8607320785522461, "learning_rate": 0.0002, "epoch": 3.196496989600438, "step": 2920}, {"loss": 0.6139, "grad_norm": 0.8854126334190369, "learning_rate": 0.0002, "epoch": 3.2074438970990693, "step": 2930}, {"loss": 0.6485, "grad_norm": 0.6620870232582092, "learning_rate": 0.0002, "epoch": 3.218390804597701, "step": 2940}, {"loss": 0.6969, "grad_norm": 0.7377511858940125, "learning_rate": 0.0002, "epoch": 3.2293377120963327, "step": 2950}, {"loss": 0.6798, "grad_norm": 0.7803301811218262, "learning_rate": 0.0002, "epoch": 3.2402846195949646, "step": 2960}, {"loss": 0.6697, "grad_norm": 0.834061861038208, "learning_rate": 0.0002, "epoch": 3.251231527093596, "step": 2970}, {"loss": 0.6894, "grad_norm": 0.8496041893959045, "learning_rate": 0.0002, "epoch": 3.2621784345922276, "step": 2980}, {"loss": 0.6591, "grad_norm": 0.7967984676361084, "learning_rate": 0.0002, "epoch": 3.2731253420908595, "step": 2990}, {"loss": 0.7266, "grad_norm": 1.0207016468048096, "learning_rate": 0.0002, "epoch": 3.284072249589491, "step": 3000}, {"loss": 0.6586, "grad_norm": 0.7855771780014038, "learning_rate": 0.0002, "epoch": 3.2950191570881224, "step": 3010}, {"loss": 0.5711, "grad_norm": 0.9427546858787537, "learning_rate": 0.0002, "epoch": 3.3059660645867543, "step": 3020}, {"loss": 0.6277, "grad_norm": 0.823542594909668, "learning_rate": 0.0002, "epoch": 3.316912972085386, "step": 3030}, {"loss": 0.7109, "grad_norm": 0.9826635122299194, "learning_rate": 0.0002, "epoch": 3.3278598795840173, "step": 3040}, {"loss": 0.6564, "grad_norm": 0.7259827852249146, "learning_rate": 0.0002, "epoch": 3.338806787082649, "step": 3050}, {"loss": 0.653, "grad_norm": 0.7774739861488342, "learning_rate": 0.0002, "epoch": 3.3497536945812807, "step": 3060}, {"loss": 0.7529, "grad_norm": 0.7394293546676636, "learning_rate": 0.0002, "epoch": 3.3607006020799126, "step": 3070}, {"loss": 0.5987, "grad_norm": 0.9017578959465027, "learning_rate": 0.0002, "epoch": 3.371647509578544, "step": 3080}, {"loss": 0.6953, "grad_norm": 0.7451054453849792, "learning_rate": 0.0002, "epoch": 3.3825944170771756, "step": 3090}, {"loss": 0.6759, "grad_norm": 0.7321506142616272, "learning_rate": 0.0002, "epoch": 3.3935413245758075, "step": 3100}, {"loss": 0.6555, "grad_norm": 0.6721828579902649, "learning_rate": 0.0002, "epoch": 3.404488232074439, "step": 3110}, {"loss": 0.6559, "grad_norm": 0.774022102355957, "learning_rate": 0.0002, "epoch": 3.4154351395730704, "step": 3120}, {"loss": 0.7449, "grad_norm": 0.9143537282943726, "learning_rate": 0.0002, "epoch": 3.4263820470717024, "step": 3130}, {"loss": 0.6899, "grad_norm": 1.226087212562561, "learning_rate": 0.0002, "epoch": 3.437328954570334, "step": 3140}, {"loss": 0.6719, "grad_norm": 0.7545496225357056, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 3150}, {"loss": 0.6153, "grad_norm": 0.6515635848045349, "learning_rate": 0.0002, "epoch": 3.4592227695675972, "step": 3160}, {"loss": 0.6926, "grad_norm": 0.9297090172767639, "learning_rate": 0.0002, "epoch": 3.4701696770662287, "step": 3170}, {"loss": 0.6071, "grad_norm": 1.0130730867385864, "learning_rate": 0.0002, "epoch": 3.4811165845648606, "step": 3180}, {"loss": 0.5959, "grad_norm": 0.7654589414596558, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 3190}, {"loss": 0.7401, "grad_norm": 0.9954977631568909, "learning_rate": 0.0002, "epoch": 3.5030103995621236, "step": 3200}, {"loss": 0.6661, "grad_norm": 0.6027487516403198, "learning_rate": 0.0002, "epoch": 3.5139573070607555, "step": 3210}, {"loss": 0.6963, "grad_norm": 0.741770327091217, "learning_rate": 0.0002, "epoch": 3.524904214559387, "step": 3220}, {"loss": 0.8112, "grad_norm": 1.0534909963607788, "learning_rate": 0.0002, "epoch": 3.535851122058019, "step": 3230}, {"loss": 0.6813, "grad_norm": 0.937772274017334, "learning_rate": 0.0002, "epoch": 3.5467980295566504, "step": 3240}, {"loss": 0.6681, "grad_norm": 0.8504213690757751, "learning_rate": 0.0002, "epoch": 3.557744937055282, "step": 3250}, {"loss": 0.6436, "grad_norm": 0.7755007147789001, "learning_rate": 0.0002, "epoch": 3.5686918445539133, "step": 3260}, {"loss": 0.6213, "grad_norm": 1.0193358659744263, "learning_rate": 0.0002, "epoch": 3.5796387520525452, "step": 3270}, {"loss": 0.671, "grad_norm": 0.8440536856651306, "learning_rate": 0.0002, "epoch": 3.5905856595511767, "step": 3280}, {"loss": 0.6859, "grad_norm": 0.6195939183235168, "learning_rate": 0.0002, "epoch": 3.6015325670498086, "step": 3290}, {"loss": 0.7446, "grad_norm": 0.8608590960502625, "learning_rate": 0.0002, "epoch": 3.61247947454844, "step": 3300}, {"loss": 0.7301, "grad_norm": 0.6772327423095703, "learning_rate": 0.0002, "epoch": 3.6234263820470716, "step": 3310}, {"loss": 0.6298, "grad_norm": 0.8031839728355408, "learning_rate": 0.0002, "epoch": 3.6343732895457035, "step": 3320}, {"loss": 0.7041, "grad_norm": 0.6080502271652222, "learning_rate": 0.0002, "epoch": 3.645320197044335, "step": 3330}, {"loss": 0.7431, "grad_norm": 0.8007240891456604, "learning_rate": 0.0002, "epoch": 3.656267104542967, "step": 3340}, {"loss": 0.7446, "grad_norm": 0.8060704469680786, "learning_rate": 0.0002, "epoch": 3.6672140120415984, "step": 3350}, {"loss": 0.6304, "grad_norm": 0.7547586560249329, "learning_rate": 0.0002, "epoch": 3.67816091954023, "step": 3360}, {"loss": 0.7066, "grad_norm": 0.686851978302002, "learning_rate": 0.0002, "epoch": 3.6891078270388613, "step": 3370}, {"loss": 0.6748, "grad_norm": 0.9429075717926025, "learning_rate": 0.0002, "epoch": 3.7000547345374932, "step": 3380}, {"loss": 0.6673, "grad_norm": 0.7283591032028198, "learning_rate": 0.0002, "epoch": 3.7110016420361247, "step": 3390}, {"loss": 0.7502, "grad_norm": 0.8323085904121399, "learning_rate": 0.0002, "epoch": 3.7219485495347566, "step": 3400}, {"loss": 0.7779, "grad_norm": 0.8529590964317322, "learning_rate": 0.0002, "epoch": 3.732895457033388, "step": 3410}, {"loss": 0.6555, "grad_norm": 0.731752872467041, "learning_rate": 0.0002, "epoch": 3.7438423645320196, "step": 3420}, {"loss": 0.6928, "grad_norm": 0.8572278618812561, "learning_rate": 0.0002, "epoch": 3.7547892720306515, "step": 3430}, {"loss": 0.6215, "grad_norm": 0.7408691048622131, "learning_rate": 0.0002, "epoch": 3.765736179529283, "step": 3440}, {"loss": 0.622, "grad_norm": 0.7470445036888123, "learning_rate": 0.0002, "epoch": 3.776683087027915, "step": 3450}, {"loss": 0.7241, "grad_norm": 0.6806244254112244, "learning_rate": 0.0002, "epoch": 3.7876299945265464, "step": 3460}, {"loss": 0.7739, "grad_norm": 0.9129069447517395, "learning_rate": 0.0002, "epoch": 3.798576902025178, "step": 3470}, {"loss": 0.6826, "grad_norm": 0.8717501759529114, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 3480}, {"loss": 0.6188, "grad_norm": 0.6761979460716248, "learning_rate": 0.0002, "epoch": 3.8204707170224412, "step": 3490}, {"loss": 0.7601, "grad_norm": 1.0054380893707275, "learning_rate": 0.0002, "epoch": 3.8314176245210727, "step": 3500}, {"loss": 0.623, "grad_norm": 1.1224009990692139, "learning_rate": 0.0002, "epoch": 3.8423645320197046, "step": 3510}, {"loss": 0.6918, "grad_norm": 0.8997692465782166, "learning_rate": 0.0002, "epoch": 3.853311439518336, "step": 3520}, {"loss": 0.6357, "grad_norm": 1.0086902379989624, "learning_rate": 0.0002, "epoch": 3.8642583470169676, "step": 3530}, {"loss": 0.6379, "grad_norm": 0.772739589214325, "learning_rate": 0.0002, "epoch": 3.8752052545155995, "step": 3540}, {"loss": 0.7423, "grad_norm": 1.211774230003357, "learning_rate": 0.0002, "epoch": 3.886152162014231, "step": 3550}, {"loss": 0.7321, "grad_norm": 0.9572356939315796, "learning_rate": 0.0002, "epoch": 3.897099069512863, "step": 3560}, {"loss": 0.6836, "grad_norm": 0.7887842655181885, "learning_rate": 0.0002, "epoch": 3.9080459770114944, "step": 3570}, {"loss": 0.7576, "grad_norm": 0.7308389544487, "learning_rate": 0.0002, "epoch": 3.918992884510126, "step": 3580}, {"loss": 0.6001, "grad_norm": 1.0182650089263916, "learning_rate": 0.0002, "epoch": 3.9299397920087573, "step": 3590}, {"loss": 0.6942, "grad_norm": 0.8000147342681885, "learning_rate": 0.0002, "epoch": 3.9408866995073892, "step": 3600}, {"loss": 0.6244, "grad_norm": 0.7385728359222412, "learning_rate": 0.0002, "epoch": 3.9518336070060207, "step": 3610}, {"loss": 0.6718, "grad_norm": 0.9233261942863464, "learning_rate": 0.0002, "epoch": 3.9627805145046526, "step": 3620}, {"loss": 0.6508, "grad_norm": 0.8486751914024353, "learning_rate": 0.0002, "epoch": 3.973727422003284, "step": 3630}, {"loss": 0.6928, "grad_norm": 0.7593663334846497, "learning_rate": 0.0002, "epoch": 3.9846743295019156, "step": 3640}, {"loss": 0.6847, "grad_norm": 0.7885415554046631, "learning_rate": 0.0002, "epoch": 3.9956212370005475, "step": 3650}, {"eval_loss": 1.250312328338623, "eval_runtime": 46.0842, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.193, "epoch": 4.0, "step": 3654}, {"loss": 0.5547, "grad_norm": 0.6591703295707703, "learning_rate": 0.0002, "epoch": 4.006568144499179, "step": 3660}, {"loss": 0.5301, "grad_norm": 1.36927330493927, "learning_rate": 0.0002, "epoch": 4.017515051997811, "step": 3670}, {"loss": 0.4466, "grad_norm": 0.8106328845024109, "learning_rate": 0.0002, "epoch": 4.028461959496442, "step": 3680}, {"loss": 0.4861, "grad_norm": 0.7592712044715881, "learning_rate": 0.0002, "epoch": 4.039408866995074, "step": 3690}, {"loss": 0.5103, "grad_norm": 0.9518909454345703, "learning_rate": 0.0002, "epoch": 4.050355774493705, "step": 3700}, {"loss": 0.4638, "grad_norm": 0.7805967330932617, "learning_rate": 0.0002, "epoch": 4.061302681992337, "step": 3710}, {"loss": 0.4556, "grad_norm": 1.3146334886550903, "learning_rate": 0.0002, "epoch": 4.072249589490969, "step": 3720}, {"loss": 0.5635, "grad_norm": 1.1611138582229614, "learning_rate": 0.0002, "epoch": 4.083196496989601, "step": 3730}, {"loss": 0.3845, "grad_norm": 0.8173232078552246, "learning_rate": 0.0002, "epoch": 4.094143404488232, "step": 3740}, {"loss": 0.4911, "grad_norm": 0.7848323583602905, "learning_rate": 0.0002, "epoch": 4.105090311986864, "step": 3750}, {"loss": 0.4519, "grad_norm": 1.3183201551437378, "learning_rate": 0.0002, "epoch": 4.116037219485495, "step": 3760}, {"loss": 0.5083, "grad_norm": 1.1936529874801636, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 3770}, {"loss": 0.5208, "grad_norm": 1.1078993082046509, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 3780}, {"loss": 0.5928, "grad_norm": 1.107743263244629, "learning_rate": 0.0002, "epoch": 4.14887794198139, "step": 3790}, {"loss": 0.5112, "grad_norm": 0.7801875472068787, "learning_rate": 0.0002, "epoch": 4.159824849480022, "step": 3800}, {"loss": 0.4896, "grad_norm": 1.1328117847442627, "learning_rate": 0.0002, "epoch": 4.170771756978653, "step": 3810}, {"loss": 0.5645, "grad_norm": 1.4232193231582642, "learning_rate": 0.0002, "epoch": 4.181718664477285, "step": 3820}, {"loss": 0.5049, "grad_norm": 1.557416558265686, "learning_rate": 0.0002, "epoch": 4.192665571975917, "step": 3830}, {"loss": 0.4863, "grad_norm": 1.042923092842102, "learning_rate": 0.0002, "epoch": 4.203612479474549, "step": 3840}, {"loss": 0.3751, "grad_norm": 1.1801949739456177, "learning_rate": 0.0002, "epoch": 4.21455938697318, "step": 3850}, {"loss": 0.5063, "grad_norm": 0.9273753762245178, "learning_rate": 0.0002, "epoch": 4.225506294471812, "step": 3860}, {"loss": 0.5542, "grad_norm": 0.7681763768196106, "learning_rate": 0.0002, "epoch": 4.236453201970443, "step": 3870}, {"loss": 0.5971, "grad_norm": 0.9840841293334961, "learning_rate": 0.0002, "epoch": 4.2474001094690745, "step": 3880}, {"loss": 0.4648, "grad_norm": 1.0290725231170654, "learning_rate": 0.0002, "epoch": 4.258347016967707, "step": 3890}, {"loss": 0.4288, "grad_norm": 0.8059597611427307, "learning_rate": 0.0002, "epoch": 4.269293924466338, "step": 3900}, {"loss": 0.5103, "grad_norm": 0.9847467541694641, "learning_rate": 0.0002, "epoch": 4.28024083196497, "step": 3910}, {"loss": 0.4952, "grad_norm": 1.344044804573059, "learning_rate": 0.0002, "epoch": 4.291187739463601, "step": 3920}, {"loss": 0.4966, "grad_norm": 0.9174224138259888, "learning_rate": 0.0002, "epoch": 4.302134646962233, "step": 3930}, {"loss": 0.4944, "grad_norm": 1.1199711561203003, "learning_rate": 0.0002, "epoch": 4.313081554460865, "step": 3940}, {"loss": 0.4641, "grad_norm": 1.0120296478271484, "learning_rate": 0.0002, "epoch": 4.324028461959497, "step": 3950}, {"loss": 0.4723, "grad_norm": 1.091811180114746, "learning_rate": 0.0002, "epoch": 4.334975369458128, "step": 3960}, {"loss": 0.4627, "grad_norm": 1.0332133769989014, "learning_rate": 0.0002, "epoch": 4.34592227695676, "step": 3970}, {"loss": 0.4646, "grad_norm": 1.0785295963287354, "learning_rate": 0.0002, "epoch": 4.356869184455391, "step": 3980}, {"loss": 0.4909, "grad_norm": 1.0506969690322876, "learning_rate": 0.0002, "epoch": 4.3678160919540225, "step": 3990}, {"loss": 0.4776, "grad_norm": 1.047560691833496, "learning_rate": 0.0002, "epoch": 4.378762999452655, "step": 4000}, {"loss": 0.4549, "grad_norm": 0.9348800778388977, "learning_rate": 0.0002, "epoch": 4.389709906951286, "step": 4010}, {"loss": 0.5333, "grad_norm": 1.1563059091567993, "learning_rate": 0.0002, "epoch": 4.400656814449918, "step": 4020}, {"loss": 0.4952, "grad_norm": 1.001470923423767, "learning_rate": 0.0002, "epoch": 4.411603721948549, "step": 4030}, {"loss": 0.4972, "grad_norm": 1.309012532234192, "learning_rate": 0.0002, "epoch": 4.422550629447181, "step": 4040}, {"loss": 0.5078, "grad_norm": 0.7338925004005432, "learning_rate": 0.0002, "epoch": 4.433497536945813, "step": 4050}, {"loss": 0.4632, "grad_norm": 1.0398834943771362, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 4060}, {"loss": 0.6285, "grad_norm": 0.9728689193725586, "learning_rate": 0.0002, "epoch": 4.455391351943076, "step": 4070}, {"loss": 0.4778, "grad_norm": 1.247475028038025, "learning_rate": 0.0002, "epoch": 4.466338259441708, "step": 4080}, {"loss": 0.4813, "grad_norm": 1.1084578037261963, "learning_rate": 0.0002, "epoch": 4.477285166940339, "step": 4090}, {"loss": 0.5665, "grad_norm": 1.1619318723678589, "learning_rate": 0.0002, "epoch": 4.4882320744389705, "step": 4100}, {"loss": 0.5207, "grad_norm": 1.3456498384475708, "learning_rate": 0.0002, "epoch": 4.499178981937603, "step": 4110}, {"loss": 0.4876, "grad_norm": 0.9372991323471069, "learning_rate": 0.0002, "epoch": 4.510125889436234, "step": 4120}, {"loss": 0.5456, "grad_norm": 1.0071815252304077, "learning_rate": 0.0002, "epoch": 4.521072796934866, "step": 4130}, {"loss": 0.5589, "grad_norm": 1.190344214439392, "learning_rate": 0.0002, "epoch": 4.532019704433497, "step": 4140}, {"loss": 0.4852, "grad_norm": 0.9480887055397034, "learning_rate": 0.0002, "epoch": 4.542966611932129, "step": 4150}, {"loss": 0.5229, "grad_norm": 1.0252189636230469, "learning_rate": 0.0002, "epoch": 4.553913519430761, "step": 4160}, {"loss": 0.5253, "grad_norm": 0.7142013311386108, "learning_rate": 0.0002, "epoch": 4.564860426929393, "step": 4170}, {"loss": 0.4861, "grad_norm": 0.8937426805496216, "learning_rate": 0.0002, "epoch": 4.575807334428024, "step": 4180}, {"loss": 0.4773, "grad_norm": 0.8885005116462708, "learning_rate": 0.0002, "epoch": 4.586754241926656, "step": 4190}, {"loss": 0.4858, "grad_norm": 1.337663173675537, "learning_rate": 0.0002, "epoch": 4.597701149425287, "step": 4200}, {"loss": 0.5247, "grad_norm": 1.0475375652313232, "learning_rate": 0.0002, "epoch": 4.6086480569239185, "step": 4210}, {"loss": 0.5298, "grad_norm": 1.0081088542938232, "learning_rate": 0.0002, "epoch": 4.619594964422551, "step": 4220}, {"loss": 0.5042, "grad_norm": 0.7527595162391663, "learning_rate": 0.0002, "epoch": 4.630541871921182, "step": 4230}, {"loss": 0.5207, "grad_norm": 1.55559241771698, "learning_rate": 0.0002, "epoch": 4.641488779419814, "step": 4240}, {"loss": 0.5468, "grad_norm": 0.7967379689216614, "learning_rate": 0.0002, "epoch": 4.652435686918445, "step": 4250}, {"loss": 0.5328, "grad_norm": 0.898368775844574, "learning_rate": 0.0002, "epoch": 4.663382594417077, "step": 4260}, {"loss": 0.4706, "grad_norm": 1.1940776109695435, "learning_rate": 0.0002, "epoch": 4.674329501915709, "step": 4270}, {"loss": 0.5121, "grad_norm": 1.1817092895507812, "learning_rate": 0.0002, "epoch": 4.685276409414341, "step": 4280}, {"loss": 0.5758, "grad_norm": 0.9041520357131958, "learning_rate": 0.0002, "epoch": 4.696223316912972, "step": 4290}, {"loss": 0.5851, "grad_norm": 1.1280102729797363, "learning_rate": 0.0002, "epoch": 4.707170224411604, "step": 4300}, {"loss": 0.4891, "grad_norm": 1.357689619064331, "learning_rate": 0.0002, "epoch": 4.718117131910235, "step": 4310}, {"loss": 0.4704, "grad_norm": 1.056633472442627, "learning_rate": 0.0002, "epoch": 4.7290640394088665, "step": 4320}, {"loss": 0.5488, "grad_norm": 1.6520427465438843, "learning_rate": 0.0002, "epoch": 4.740010946907499, "step": 4330}, {"loss": 0.5131, "grad_norm": 1.153200626373291, "learning_rate": 0.0002, "epoch": 4.75095785440613, "step": 4340}, {"loss": 0.539, "grad_norm": 0.9346241354942322, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 4350}, {"loss": 0.4941, "grad_norm": 0.8628455996513367, "learning_rate": 0.0002, "epoch": 4.772851669403393, "step": 4360}, {"loss": 0.5167, "grad_norm": 1.3843916654586792, "learning_rate": 0.0002, "epoch": 4.783798576902025, "step": 4370}, {"loss": 0.4683, "grad_norm": 1.035574197769165, "learning_rate": 0.0002, "epoch": 4.794745484400657, "step": 4380}, {"loss": 0.5162, "grad_norm": 1.1868361234664917, "learning_rate": 0.0002, "epoch": 4.805692391899289, "step": 4390}, {"loss": 0.534, "grad_norm": 1.1307647228240967, "learning_rate": 0.0002, "epoch": 4.81663929939792, "step": 4400}, {"loss": 0.5567, "grad_norm": 0.9787724614143372, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 4410}, {"loss": 0.5185, "grad_norm": 1.0473824739456177, "learning_rate": 0.0002, "epoch": 4.838533114395183, "step": 4420}, {"loss": 0.6285, "grad_norm": 1.069069504737854, "learning_rate": 0.0002, "epoch": 4.8494800218938146, "step": 4430}, {"loss": 0.5267, "grad_norm": 1.4305680990219116, "learning_rate": 0.0002, "epoch": 4.860426929392447, "step": 4440}, {"loss": 0.5947, "grad_norm": 1.3679203987121582, "learning_rate": 0.0002, "epoch": 4.871373836891078, "step": 4450}, {"loss": 0.5135, "grad_norm": 0.8997844457626343, "learning_rate": 0.0002, "epoch": 4.88232074438971, "step": 4460}, {"loss": 0.5312, "grad_norm": 1.2758110761642456, "learning_rate": 0.0002, "epoch": 4.893267651888341, "step": 4470}, {"loss": 0.4914, "grad_norm": 0.8819465637207031, "learning_rate": 0.0002, "epoch": 4.904214559386973, "step": 4480}, {"loss": 0.5147, "grad_norm": 1.08329439163208, "learning_rate": 0.0002, "epoch": 4.915161466885605, "step": 4490}, {"loss": 0.5404, "grad_norm": 1.083461046218872, "learning_rate": 0.0002, "epoch": 4.926108374384237, "step": 4500}, {"loss": 0.5433, "grad_norm": 1.2387723922729492, "learning_rate": 0.0002, "epoch": 4.937055281882868, "step": 4510}, {"loss": 0.5624, "grad_norm": 0.8262293934822083, "learning_rate": 0.0002, "epoch": 4.9480021893815, "step": 4520}, {"loss": 0.504, "grad_norm": 1.2325191497802734, "learning_rate": 0.0002, "epoch": 4.958949096880131, "step": 4530}, {"loss": 0.5452, "grad_norm": 1.024614930152893, "learning_rate": 0.0002, "epoch": 4.9698960043787626, "step": 4540}, {"loss": 0.4752, "grad_norm": 1.3007521629333496, "learning_rate": 0.0002, "epoch": 4.980842911877395, "step": 4550}, {"loss": 0.4943, "grad_norm": 0.9823828339576721, "learning_rate": 0.0002, "epoch": 4.991789819376026, "step": 4560}, {"eval_loss": 1.3920727968215942, "eval_runtime": 46.0764, "eval_samples_per_second": 9.463, "eval_steps_per_second": 1.194, "epoch": 4.999452654625069, "step": 4567}, {"loss": 0.545, "grad_norm": 1.1478906869888306, "learning_rate": 0.0002, "epoch": 5.002736726874658, "step": 4570}, {"loss": 0.372, "grad_norm": 1.0533705949783325, "learning_rate": 0.0002, "epoch": 5.013683634373289, "step": 4580}, {"loss": 0.3313, "grad_norm": 1.268900752067566, "learning_rate": 0.0002, "epoch": 5.024630541871921, "step": 4590}, {"loss": 0.3482, "grad_norm": 1.222652554512024, "learning_rate": 0.0002, "epoch": 5.035577449370553, "step": 4600}, {"loss": 0.3195, "grad_norm": 1.5093127489089966, "learning_rate": 0.0002, "epoch": 5.046524356869185, "step": 4610}, {"loss": 0.3569, "grad_norm": 1.2372499704360962, "learning_rate": 0.0002, "epoch": 5.057471264367816, "step": 4620}, {"loss": 0.3206, "grad_norm": 0.8422666192054749, "learning_rate": 0.0002, "epoch": 5.068418171866448, "step": 4630}, {"loss": 0.3115, "grad_norm": 1.1451770067214966, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 4640}, {"loss": 0.3305, "grad_norm": 1.2074557542800903, "learning_rate": 0.0002, "epoch": 5.090311986863711, "step": 4650}, {"loss": 0.3012, "grad_norm": 1.429150104522705, "learning_rate": 0.0002, "epoch": 5.101258894362343, "step": 4660}, {"loss": 0.3229, "grad_norm": 1.0353610515594482, "learning_rate": 0.0002, "epoch": 5.112205801860974, "step": 4670}, {"loss": 0.402, "grad_norm": 1.2845979928970337, "learning_rate": 0.0002, "epoch": 5.123152709359606, "step": 4680}, {"loss": 0.383, "grad_norm": 1.3790186643600464, "learning_rate": 0.0002, "epoch": 5.134099616858237, "step": 4690}, {"loss": 0.2951, "grad_norm": 1.3182239532470703, "learning_rate": 0.0002, "epoch": 5.145046524356869, "step": 4700}, {"loss": 0.4074, "grad_norm": 1.5249626636505127, "learning_rate": 0.0002, "epoch": 5.155993431855501, "step": 4710}, {"loss": 0.3703, "grad_norm": 1.2492733001708984, "learning_rate": 0.0002, "epoch": 5.166940339354133, "step": 4720}, {"loss": 0.3411, "grad_norm": 1.4455480575561523, "learning_rate": 0.0002, "epoch": 5.177887246852764, "step": 4730}, {"loss": 0.3996, "grad_norm": 1.2191482782363892, "learning_rate": 0.0002, "epoch": 5.188834154351396, "step": 4740}, {"loss": 0.3785, "grad_norm": 1.4707951545715332, "learning_rate": 0.0002, "epoch": 5.199781061850027, "step": 4750}, {"loss": 0.3516, "grad_norm": 1.3473678827285767, "learning_rate": 0.0002, "epoch": 5.210727969348659, "step": 4760}, {"loss": 0.3266, "grad_norm": 1.0479670763015747, "learning_rate": 0.0002, "epoch": 5.221674876847291, "step": 4770}, {"loss": 0.3976, "grad_norm": 1.299096703529358, "learning_rate": 0.0002, "epoch": 5.232621784345922, "step": 4780}, {"loss": 0.3266, "grad_norm": 1.2820168733596802, "learning_rate": 0.0002, "epoch": 5.243568691844554, "step": 4790}, {"loss": 0.3347, "grad_norm": 1.3818004131317139, "learning_rate": 0.0002, "epoch": 5.254515599343185, "step": 4800}, {"loss": 0.3761, "grad_norm": 1.2898736000061035, "learning_rate": 0.0002, "epoch": 5.265462506841817, "step": 4810}, {"loss": 0.3694, "grad_norm": 1.1761468648910522, "learning_rate": 0.0002, "epoch": 5.276409414340449, "step": 4820}, {"loss": 0.3806, "grad_norm": 1.7155952453613281, "learning_rate": 0.0002, "epoch": 5.287356321839081, "step": 4830}, {"loss": 0.322, "grad_norm": 0.9103642106056213, "learning_rate": 0.0002, "epoch": 5.298303229337712, "step": 4840}, {"loss": 0.3516, "grad_norm": 1.013015627861023, "learning_rate": 0.0002, "epoch": 5.309250136836344, "step": 4850}, {"loss": 0.4297, "grad_norm": 1.390471339225769, "learning_rate": 0.0002, "epoch": 5.320197044334975, "step": 4860}, {"loss": 0.4098, "grad_norm": 1.129770278930664, "learning_rate": 0.0002, "epoch": 5.331143951833607, "step": 4870}, {"loss": 0.4227, "grad_norm": 1.1461067199707031, "learning_rate": 0.0002, "epoch": 5.342090859332239, "step": 4880}, {"loss": 0.288, "grad_norm": 1.3587424755096436, "learning_rate": 0.0002, "epoch": 5.35303776683087, "step": 4890}, {"loss": 0.3604, "grad_norm": 1.6897879838943481, "learning_rate": 0.0002, "epoch": 5.363984674329502, "step": 4900}, {"loss": 0.3887, "grad_norm": 0.9298055768013, "learning_rate": 0.0002, "epoch": 5.374931581828133, "step": 4910}, {"loss": 0.3371, "grad_norm": 1.0006917715072632, "learning_rate": 0.0002, "epoch": 5.385878489326765, "step": 4920}, {"loss": 0.3992, "grad_norm": 1.232581377029419, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 4930}, {"loss": 0.3456, "grad_norm": 1.0822620391845703, "learning_rate": 0.0002, "epoch": 5.407772304324029, "step": 4940}, {"loss": 0.3806, "grad_norm": 1.3648720979690552, "learning_rate": 0.0002, "epoch": 5.41871921182266, "step": 4950}, {"loss": 0.3959, "grad_norm": 1.3220354318618774, "learning_rate": 0.0002, "epoch": 5.429666119321292, "step": 4960}, {"loss": 0.3278, "grad_norm": 1.1106271743774414, "learning_rate": 0.0002, "epoch": 5.440613026819923, "step": 4970}, {"loss": 0.3812, "grad_norm": 1.6058908700942993, "learning_rate": 0.0002, "epoch": 5.451559934318555, "step": 4980}, {"loss": 0.3905, "grad_norm": 1.1065930128097534, "learning_rate": 0.0002, "epoch": 5.462506841817187, "step": 4990}, {"loss": 0.4058, "grad_norm": 1.3896466493606567, "learning_rate": 0.0002, "epoch": 5.473453749315818, "step": 5000}, {"loss": 0.4122, "grad_norm": 1.0437148809432983, "learning_rate": 0.0002, "epoch": 5.48440065681445, "step": 5010}, {"loss": 0.4065, "grad_norm": 1.2347718477249146, "learning_rate": 0.0002, "epoch": 5.495347564313081, "step": 5020}, {"loss": 0.3586, "grad_norm": 1.1174284219741821, "learning_rate": 0.0002, "epoch": 5.506294471811713, "step": 5030}, {"loss": 0.3576, "grad_norm": 1.2580941915512085, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 5040}, {"loss": 0.3809, "grad_norm": 1.451090931892395, "learning_rate": 0.0002, "epoch": 5.528188286808977, "step": 5050}, {"loss": 0.3645, "grad_norm": 1.4688365459442139, "learning_rate": 0.0002, "epoch": 5.539135194307608, "step": 5060}, {"loss": 0.4431, "grad_norm": 1.1625734567642212, "learning_rate": 0.0002, "epoch": 5.55008210180624, "step": 5070}, {"loss": 0.3972, "grad_norm": 0.9332265257835388, "learning_rate": 0.0002, "epoch": 5.561029009304871, "step": 5080}, {"loss": 0.4, "grad_norm": 1.5635273456573486, "learning_rate": 0.0002, "epoch": 5.571975916803503, "step": 5090}, {"loss": 0.3651, "grad_norm": 1.3420509099960327, "learning_rate": 0.0002, "epoch": 5.582922824302135, "step": 5100}, {"loss": 0.3717, "grad_norm": 1.5826557874679565, "learning_rate": 0.0002, "epoch": 5.593869731800766, "step": 5110}, {"loss": 0.4256, "grad_norm": 1.5737065076828003, "learning_rate": 0.0002, "epoch": 5.604816639299398, "step": 5120}, {"loss": 0.39, "grad_norm": 1.3812499046325684, "learning_rate": 0.0002, "epoch": 5.615763546798029, "step": 5130}, {"loss": 0.3891, "grad_norm": 1.362833023071289, "learning_rate": 0.0002, "epoch": 5.626710454296661, "step": 5140}, {"loss": 0.455, "grad_norm": 1.7667874097824097, "learning_rate": 0.0002, "epoch": 5.637657361795293, "step": 5150}, {"loss": 0.4264, "grad_norm": 1.2661789655685425, "learning_rate": 0.0002, "epoch": 5.648604269293925, "step": 5160}, {"loss": 0.3261, "grad_norm": 1.2076870203018188, "learning_rate": 0.0002, "epoch": 5.659551176792556, "step": 5170}, {"loss": 0.372, "grad_norm": 1.2431524991989136, "learning_rate": 0.0002, "epoch": 5.670498084291188, "step": 5180}, {"loss": 0.4092, "grad_norm": 1.2216639518737793, "learning_rate": 0.0002, "epoch": 5.681444991789819, "step": 5190}, {"loss": 0.4171, "grad_norm": 0.9259352684020996, "learning_rate": 0.0002, "epoch": 5.692391899288451, "step": 5200}, {"loss": 0.3875, "grad_norm": 1.7929338216781616, "learning_rate": 0.0002, "epoch": 5.703338806787083, "step": 5210}, {"loss": 0.4424, "grad_norm": 1.4048460721969604, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 5220}, {"loss": 0.3758, "grad_norm": 1.306874394416809, "learning_rate": 0.0002, "epoch": 5.725232621784346, "step": 5230}, {"loss": 0.3889, "grad_norm": 1.3137940168380737, "learning_rate": 0.0002, "epoch": 5.736179529282977, "step": 5240}, {"loss": 0.4804, "grad_norm": 1.1376476287841797, "learning_rate": 0.0002, "epoch": 5.747126436781609, "step": 5250}, {"loss": 0.377, "grad_norm": 1.450939416885376, "learning_rate": 0.0002, "epoch": 5.758073344280241, "step": 5260}, {"loss": 0.4732, "grad_norm": 0.983195960521698, "learning_rate": 0.0002, "epoch": 5.769020251778873, "step": 5270}, {"loss": 0.4041, "grad_norm": 1.66558837890625, "learning_rate": 0.0002, "epoch": 5.779967159277504, "step": 5280}, {"loss": 0.3643, "grad_norm": 0.9789204597473145, "learning_rate": 0.0002, "epoch": 5.790914066776136, "step": 5290}, {"loss": 0.3776, "grad_norm": 1.2110556364059448, "learning_rate": 0.0002, "epoch": 5.801860974274767, "step": 5300}, {"loss": 0.4049, "grad_norm": 1.3799304962158203, "learning_rate": 0.0002, "epoch": 5.812807881773399, "step": 5310}, {"loss": 0.4362, "grad_norm": 1.0570626258850098, "learning_rate": 0.0002, "epoch": 5.823754789272031, "step": 5320}, {"loss": 0.4716, "grad_norm": 1.4654436111450195, "learning_rate": 0.0002, "epoch": 5.834701696770662, "step": 5330}, {"loss": 0.4048, "grad_norm": 1.5216940641403198, "learning_rate": 0.0002, "epoch": 5.845648604269294, "step": 5340}, {"loss": 0.3848, "grad_norm": 1.018646001815796, "learning_rate": 0.0002, "epoch": 5.856595511767925, "step": 5350}, {"loss": 0.3705, "grad_norm": 1.028951644897461, "learning_rate": 0.0002, "epoch": 5.867542419266557, "step": 5360}, {"loss": 0.4213, "grad_norm": 2.571263313293457, "learning_rate": 0.0002, "epoch": 5.878489326765189, "step": 5370}, {"loss": 0.3647, "grad_norm": 1.3323984146118164, "learning_rate": 0.0002, "epoch": 5.889436234263821, "step": 5380}, {"loss": 0.4085, "grad_norm": 1.4317777156829834, "learning_rate": 0.0002, "epoch": 5.900383141762452, "step": 5390}, {"loss": 0.4254, "grad_norm": 1.4289140701293945, "learning_rate": 0.0002, "epoch": 5.911330049261084, "step": 5400}, {"loss": 0.3993, "grad_norm": 1.3130780458450317, "learning_rate": 0.0002, "epoch": 5.922276956759715, "step": 5410}, {"loss": 0.4025, "grad_norm": 1.3979902267456055, "learning_rate": 0.0002, "epoch": 5.933223864258347, "step": 5420}, {"loss": 0.3997, "grad_norm": 1.1827352046966553, "learning_rate": 0.0002, "epoch": 5.944170771756979, "step": 5430}, {"loss": 0.4163, "grad_norm": 1.1672080755233765, "learning_rate": 0.0002, "epoch": 5.95511767925561, "step": 5440}, {"loss": 0.4425, "grad_norm": 1.0949620008468628, "learning_rate": 0.0002, "epoch": 5.966064586754242, "step": 5450}, {"loss": 0.4219, "grad_norm": 1.3183925151824951, "learning_rate": 0.0002, "epoch": 5.977011494252873, "step": 5460}, {"loss": 0.4171, "grad_norm": 1.096198320388794, "learning_rate": 0.0002, "epoch": 5.987958401751505, "step": 5470}, {"loss": 0.3886, "grad_norm": 1.2601423263549805, "learning_rate": 0.0002, "epoch": 5.998905309250137, "step": 5480}, {"eval_loss": 1.611358880996704, "eval_runtime": 46.0638, "eval_samples_per_second": 9.465, "eval_steps_per_second": 1.194, "epoch": 6.0, "step": 5481}, {"loss": 0.2616, "grad_norm": 0.9854364991188049, "learning_rate": 0.0002, "epoch": 6.009852216748769, "step": 5490}, {"loss": 0.2412, "grad_norm": 1.8073689937591553, "learning_rate": 0.0002, "epoch": 6.0207991242474, "step": 5500}, {"loss": 0.2317, "grad_norm": 1.1852164268493652, "learning_rate": 0.0002, "epoch": 6.031746031746032, "step": 5510}, {"loss": 0.224, "grad_norm": 1.0937914848327637, "learning_rate": 0.0002, "epoch": 6.042692939244663, "step": 5520}, {"loss": 0.2473, "grad_norm": 0.7411194443702698, "learning_rate": 0.0002, "epoch": 6.053639846743295, "step": 5530}, {"loss": 0.2846, "grad_norm": 1.552127480506897, "learning_rate": 0.0002, "epoch": 6.064586754241927, "step": 5540}, {"loss": 0.2639, "grad_norm": 1.0465604066848755, "learning_rate": 0.0002, "epoch": 6.075533661740558, "step": 5550}, {"loss": 0.2696, "grad_norm": 1.4008121490478516, "learning_rate": 0.0002, "epoch": 6.08648056923919, "step": 5560}, {"loss": 0.3049, "grad_norm": 1.7049046754837036, "learning_rate": 0.0002, "epoch": 6.097427476737821, "step": 5570}, {"loss": 0.263, "grad_norm": 1.111151933670044, "learning_rate": 0.0002, "epoch": 6.108374384236453, "step": 5580}, {"loss": 0.2816, "grad_norm": 1.4271087646484375, "learning_rate": 0.0002, "epoch": 6.119321291735085, "step": 5590}, {"loss": 0.2878, "grad_norm": 1.3917373418807983, "learning_rate": 0.0002, "epoch": 6.130268199233717, "step": 5600}, {"loss": 0.2482, "grad_norm": 1.013689637184143, "learning_rate": 0.0002, "epoch": 6.141215106732348, "step": 5610}, {"loss": 0.2841, "grad_norm": 1.342645525932312, "learning_rate": 0.0002, "epoch": 6.15216201423098, "step": 5620}, {"loss": 0.2335, "grad_norm": 1.4480562210083008, "learning_rate": 0.0002, "epoch": 6.163108921729611, "step": 5630}, {"loss": 0.2696, "grad_norm": 1.2483175992965698, "learning_rate": 0.0002, "epoch": 6.174055829228243, "step": 5640}, {"loss": 0.2656, "grad_norm": 1.2944550514221191, "learning_rate": 0.0002, "epoch": 6.185002736726875, "step": 5650}, {"loss": 0.2704, "grad_norm": 1.264142632484436, "learning_rate": 0.0002, "epoch": 6.195949644225506, "step": 5660}, {"loss": 0.2971, "grad_norm": 1.2068781852722168, "learning_rate": 0.0002, "epoch": 6.206896551724138, "step": 5670}, {"loss": 0.2882, "grad_norm": 1.0401629209518433, "learning_rate": 0.0002, "epoch": 6.217843459222769, "step": 5680}, {"loss": 0.3022, "grad_norm": 1.2054402828216553, "learning_rate": 0.0002, "epoch": 6.228790366721401, "step": 5690}, {"loss": 0.2949, "grad_norm": 1.1278687715530396, "learning_rate": 0.0002, "epoch": 6.239737274220033, "step": 5700}, {"loss": 0.2477, "grad_norm": 1.24592125415802, "learning_rate": 0.0002, "epoch": 6.250684181718665, "step": 5710}, {"loss": 0.246, "grad_norm": 1.2686697244644165, "learning_rate": 0.0002, "epoch": 6.261631089217296, "step": 5720}, {"loss": 0.2974, "grad_norm": 1.1836518049240112, "learning_rate": 0.0002, "epoch": 6.272577996715928, "step": 5730}, {"loss": 0.2963, "grad_norm": 1.387752890586853, "learning_rate": 0.0002, "epoch": 6.283524904214559, "step": 5740}, {"loss": 0.2961, "grad_norm": 1.9390363693237305, "learning_rate": 0.0002, "epoch": 6.294471811713191, "step": 5750}, {"loss": 0.2765, "grad_norm": 1.2919824123382568, "learning_rate": 0.0002, "epoch": 6.305418719211823, "step": 5760}, {"loss": 0.2898, "grad_norm": 1.2793965339660645, "learning_rate": 0.0002, "epoch": 6.316365626710454, "step": 5770}, {"loss": 0.2786, "grad_norm": 1.5486980676651, "learning_rate": 0.0002, "epoch": 6.327312534209086, "step": 5780}, {"loss": 0.2684, "grad_norm": 1.2757408618927002, "learning_rate": 0.0002, "epoch": 6.338259441707717, "step": 5790}, {"loss": 0.2841, "grad_norm": 1.3245713710784912, "learning_rate": 0.0002, "epoch": 6.349206349206349, "step": 5800}, {"loss": 0.3096, "grad_norm": 1.6262527704238892, "learning_rate": 0.0002, "epoch": 6.360153256704981, "step": 5810}, {"loss": 0.3219, "grad_norm": 1.465224027633667, "learning_rate": 0.0002, "epoch": 6.371100164203613, "step": 5820}, {"loss": 0.2703, "grad_norm": 1.437408447265625, "learning_rate": 0.0002, "epoch": 6.382047071702244, "step": 5830}, {"loss": 0.3012, "grad_norm": 1.3094626665115356, "learning_rate": 0.0002, "epoch": 6.392993979200876, "step": 5840}, {"loss": 0.2991, "grad_norm": 1.6717544794082642, "learning_rate": 0.0002, "epoch": 6.403940886699507, "step": 5850}, {"loss": 0.2892, "grad_norm": 1.1023344993591309, "learning_rate": 0.0002, "epoch": 6.414887794198139, "step": 5860}, {"loss": 0.3078, "grad_norm": 1.2397106885910034, "learning_rate": 0.0002, "epoch": 6.425834701696771, "step": 5870}, {"loss": 0.2984, "grad_norm": 1.6139185428619385, "learning_rate": 0.0002, "epoch": 6.436781609195402, "step": 5880}, {"loss": 0.2353, "grad_norm": 1.3164576292037964, "learning_rate": 0.0002, "epoch": 6.447728516694034, "step": 5890}, {"loss": 0.2772, "grad_norm": 1.3317217826843262, "learning_rate": 0.0002, "epoch": 6.458675424192665, "step": 5900}, {"loss": 0.2555, "grad_norm": 1.215008020401001, "learning_rate": 0.0002, "epoch": 6.469622331691297, "step": 5910}, {"loss": 0.2715, "grad_norm": 1.625672698020935, "learning_rate": 0.0002, "epoch": 6.480569239189929, "step": 5920}, {"loss": 0.2938, "grad_norm": 1.1262489557266235, "learning_rate": 0.0002, "epoch": 6.491516146688561, "step": 5930}, {"loss": 0.2921, "grad_norm": 1.447100281715393, "learning_rate": 0.0002, "epoch": 6.502463054187192, "step": 5940}, {"loss": 0.3059, "grad_norm": 1.3306448459625244, "learning_rate": 0.0002, "epoch": 6.513409961685824, "step": 5950}, {"loss": 0.2922, "grad_norm": 1.307732105255127, "learning_rate": 0.0002, "epoch": 6.524356869184455, "step": 5960}, {"loss": 0.2891, "grad_norm": 1.1851097345352173, "learning_rate": 0.0002, "epoch": 6.535303776683087, "step": 5970}, {"loss": 0.2859, "grad_norm": 1.462816596031189, "learning_rate": 0.0002, "epoch": 6.546250684181719, "step": 5980}, {"loss": 0.2698, "grad_norm": 1.2324728965759277, "learning_rate": 0.0002, "epoch": 6.55719759168035, "step": 5990}, {"loss": 0.2672, "grad_norm": 1.3627429008483887, "learning_rate": 0.0002, "epoch": 6.568144499178982, "step": 6000}, {"loss": 0.3182, "grad_norm": 1.94977867603302, "learning_rate": 0.0002, "epoch": 6.579091406677613, "step": 6010}, {"loss": 0.3183, "grad_norm": 1.459844946861267, "learning_rate": 0.0002, "epoch": 6.590038314176245, "step": 6020}, {"loss": 0.3142, "grad_norm": 1.4454325437545776, "learning_rate": 0.0002, "epoch": 6.600985221674877, "step": 6030}, {"loss": 0.269, "grad_norm": 1.4245165586471558, "learning_rate": 0.0002, "epoch": 6.611932129173509, "step": 6040}, {"loss": 0.3041, "grad_norm": 1.195803165435791, "learning_rate": 0.0002, "epoch": 6.62287903667214, "step": 6050}, {"loss": 0.3075, "grad_norm": 1.3589898347854614, "learning_rate": 0.0002, "epoch": 6.633825944170772, "step": 6060}, {"loss": 0.3291, "grad_norm": 1.3488036394119263, "learning_rate": 0.0002, "epoch": 6.644772851669403, "step": 6070}, {"loss": 0.2898, "grad_norm": 1.0954102277755737, "learning_rate": 0.0002, "epoch": 6.655719759168035, "step": 6080}, {"loss": 0.3489, "grad_norm": 1.4431062936782837, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 6090}, {"loss": 0.2816, "grad_norm": 1.4387465715408325, "learning_rate": 0.0002, "epoch": 6.677613574165298, "step": 6100}, {"loss": 0.2705, "grad_norm": 1.8398990631103516, "learning_rate": 0.0002, "epoch": 6.68856048166393, "step": 6110}, {"loss": 0.3214, "grad_norm": 1.3523273468017578, "learning_rate": 0.0002, "epoch": 6.699507389162561, "step": 6120}, {"loss": 0.287, "grad_norm": 1.6326191425323486, "learning_rate": 0.0002, "epoch": 6.710454296661193, "step": 6130}, {"loss": 0.2857, "grad_norm": 1.3677960634231567, "learning_rate": 0.0002, "epoch": 6.721401204159825, "step": 6140}, {"loss": 0.3264, "grad_norm": 1.1993201971054077, "learning_rate": 0.0002, "epoch": 6.732348111658457, "step": 6150}, {"loss": 0.3071, "grad_norm": 1.1864078044891357, "learning_rate": 0.0002, "epoch": 6.743295019157088, "step": 6160}, {"loss": 0.3087, "grad_norm": 1.1625522375106812, "learning_rate": 0.0002, "epoch": 6.75424192665572, "step": 6170}, {"loss": 0.3551, "grad_norm": 1.5803234577178955, "learning_rate": 0.0002, "epoch": 6.765188834154351, "step": 6180}, {"loss": 0.3059, "grad_norm": 1.151746153831482, "learning_rate": 0.0002, "epoch": 6.776135741652983, "step": 6190}, {"loss": 0.2697, "grad_norm": 1.0727161169052124, "learning_rate": 0.0002, "epoch": 6.787082649151615, "step": 6200}, {"loss": 0.2844, "grad_norm": 1.4148162603378296, "learning_rate": 0.0002, "epoch": 6.798029556650246, "step": 6210}, {"loss": 0.3417, "grad_norm": 1.2071447372436523, "learning_rate": 0.0002, "epoch": 6.808976464148878, "step": 6220}, {"loss": 0.3066, "grad_norm": 1.3843804597854614, "learning_rate": 0.0002, "epoch": 6.819923371647509, "step": 6230}, {"loss": 0.2769, "grad_norm": 1.2490662336349487, "learning_rate": 0.0002, "epoch": 6.830870279146141, "step": 6240}, {"loss": 0.3237, "grad_norm": 1.6029689311981201, "learning_rate": 0.0002, "epoch": 6.841817186644773, "step": 6250}, {"loss": 0.3152, "grad_norm": 1.0388455390930176, "learning_rate": 0.0002, "epoch": 6.852764094143405, "step": 6260}, {"loss": 0.3026, "grad_norm": 1.3883857727050781, "learning_rate": 0.0002, "epoch": 6.863711001642036, "step": 6270}, {"loss": 0.3175, "grad_norm": 1.0500187873840332, "learning_rate": 0.0002, "epoch": 6.874657909140668, "step": 6280}, {"loss": 0.2952, "grad_norm": 1.4243487119674683, "learning_rate": 0.0002, "epoch": 6.885604816639299, "step": 6290}, {"loss": 0.2679, "grad_norm": 1.3169665336608887, "learning_rate": 0.0002, "epoch": 6.896551724137931, "step": 6300}, {"loss": 0.3291, "grad_norm": 1.5261493921279907, "learning_rate": 0.0002, "epoch": 6.907498631636563, "step": 6310}, {"loss": 0.3344, "grad_norm": 1.578403115272522, "learning_rate": 0.0002, "epoch": 6.9184455391351944, "step": 6320}, {"loss": 0.3263, "grad_norm": 1.4093263149261475, "learning_rate": 0.0002, "epoch": 6.929392446633826, "step": 6330}, {"loss": 0.3396, "grad_norm": 1.4003552198410034, "learning_rate": 0.0002, "epoch": 6.940339354132457, "step": 6340}, {"loss": 0.3476, "grad_norm": 1.650190830230713, "learning_rate": 0.0002, "epoch": 6.951286261631089, "step": 6350}, {"loss": 0.3442, "grad_norm": 1.2314515113830566, "learning_rate": 0.0002, "epoch": 6.962233169129721, "step": 6360}, {"loss": 0.3341, "grad_norm": 1.270980954170227, "learning_rate": 0.0002, "epoch": 6.973180076628353, "step": 6370}, {"loss": 0.3425, "grad_norm": 1.6352545022964478, "learning_rate": 0.0002, "epoch": 6.984126984126984, "step": 6380}, {"loss": 0.3647, "grad_norm": 1.3744925260543823, "learning_rate": 0.0002, "epoch": 6.995073891625616, "step": 6390}]}
+{"epoch": 7.995621237000547, "step": 7304, "epoch_duration": 1353.653870344162, "total_accumulated_duration": 10884.245246648788, "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": 28746.0}, "peak_memory_reserved": {"GPU_0": 28746.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-5885-sd-4/checkpoint-1827", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4039, "grad_norm": 0.7611560821533203, "learning_rate": 0.0002, "epoch": 0.010946907498631636, "step": 10}, {"loss": 1.8818, "grad_norm": 0.4633193612098694, "learning_rate": 0.0002, "epoch": 0.021893814997263273, "step": 20}, {"loss": 1.5927, "grad_norm": 0.49326154589653015, "learning_rate": 0.0002, "epoch": 0.03284072249589491, "step": 30}, {"loss": 1.3859, "grad_norm": 0.3943138122558594, "learning_rate": 0.0002, "epoch": 0.043787629994526546, "step": 40}, {"loss": 1.3036, "grad_norm": 0.43292930722236633, "learning_rate": 0.0002, "epoch": 0.05473453749315818, "step": 50}, {"loss": 1.2427, "grad_norm": 0.3431817591190338, "learning_rate": 0.0002, "epoch": 0.06568144499178982, "step": 60}, {"loss": 1.3076, "grad_norm": 0.38011446595191956, "learning_rate": 0.0002, "epoch": 0.07662835249042145, "step": 70}, {"loss": 1.1641, "grad_norm": 0.7441071271896362, "learning_rate": 0.0002, "epoch": 0.08757525998905309, "step": 80}, {"loss": 1.1721, "grad_norm": 0.3359833061695099, "learning_rate": 0.0002, "epoch": 0.09852216748768473, "step": 90}, {"loss": 1.2774, "grad_norm": 0.3724392354488373, "learning_rate": 0.0002, "epoch": 0.10946907498631636, "step": 100}, {"loss": 1.216, "grad_norm": 0.40673762559890747, "learning_rate": 0.0002, "epoch": 0.120415982484948, "step": 110}, {"loss": 1.2611, "grad_norm": 0.40036800503730774, "learning_rate": 0.0002, "epoch": 0.13136288998357964, "step": 120}, {"loss": 1.2436, "grad_norm": 2.844191312789917, "learning_rate": 0.0002, "epoch": 0.1423097974822113, "step": 130}, {"loss": 1.2254, "grad_norm": 0.3104734420776367, "learning_rate": 0.0002, "epoch": 0.1532567049808429, "step": 140}, {"loss": 1.0498, "grad_norm": 0.3266797959804535, "learning_rate": 0.0002, "epoch": 0.16420361247947454, "step": 150}, {"loss": 1.1644, "grad_norm": 0.3079199194908142, "learning_rate": 0.0002, "epoch": 0.17515051997810618, "step": 160}, {"loss": 1.2923, "grad_norm": 0.3872479498386383, "learning_rate": 0.0002, "epoch": 0.18609742747673783, "step": 170}, {"loss": 1.1809, "grad_norm": 0.38654500246047974, "learning_rate": 0.0002, "epoch": 0.19704433497536947, "step": 180}, {"loss": 1.066, "grad_norm": 0.2913552522659302, "learning_rate": 0.0002, "epoch": 0.20799124247400108, "step": 190}, {"loss": 1.0868, "grad_norm": 0.2960572838783264, "learning_rate": 0.0002, "epoch": 0.21893814997263272, "step": 200}, {"loss": 1.136, "grad_norm": 0.5175501108169556, "learning_rate": 0.0002, "epoch": 0.22988505747126436, "step": 210}, {"loss": 1.1556, "grad_norm": 1.2921574115753174, "learning_rate": 0.0002, "epoch": 0.240831964969896, "step": 220}, {"loss": 1.1488, "grad_norm": 0.2675512135028839, "learning_rate": 0.0002, "epoch": 0.25177887246852765, "step": 230}, {"loss": 1.2764, "grad_norm": 0.3956190049648285, "learning_rate": 0.0002, "epoch": 0.2627257799671593, "step": 240}, {"loss": 1.1889, "grad_norm": 0.6022581458091736, "learning_rate": 0.0002, "epoch": 0.27367268746579093, "step": 250}, {"loss": 1.1981, "grad_norm": 1.1949563026428223, "learning_rate": 0.0002, "epoch": 0.2846195949644226, "step": 260}, {"loss": 1.1877, "grad_norm": 0.31173548102378845, "learning_rate": 0.0002, "epoch": 0.2955665024630542, "step": 270}, {"loss": 1.06, "grad_norm": 0.2808472812175751, "learning_rate": 0.0002, "epoch": 0.3065134099616858, "step": 280}, {"loss": 1.1752, "grad_norm": 0.3042023777961731, "learning_rate": 0.0002, "epoch": 0.31746031746031744, "step": 290}, {"loss": 1.3147, "grad_norm": 0.39915043115615845, "learning_rate": 0.0002, "epoch": 0.3284072249589491, "step": 300}, {"loss": 1.2425, "grad_norm": 0.39118197560310364, "learning_rate": 0.0002, "epoch": 0.3393541324575807, "step": 310}, {"loss": 1.1363, "grad_norm": 0.355010986328125, "learning_rate": 0.0002, "epoch": 0.35030103995621237, "step": 320}, {"loss": 1.1925, "grad_norm": 0.29734086990356445, "learning_rate": 0.0002, "epoch": 0.361247947454844, "step": 330}, {"loss": 1.1974, "grad_norm": 0.346096009016037, "learning_rate": 0.0002, "epoch": 0.37219485495347565, "step": 340}, {"loss": 1.1641, "grad_norm": 0.4829643666744232, "learning_rate": 0.0002, "epoch": 0.3831417624521073, "step": 350}, {"loss": 1.2808, "grad_norm": 0.4726872742176056, "learning_rate": 0.0002, "epoch": 0.39408866995073893, "step": 360}, {"loss": 1.1532, "grad_norm": 0.3130153715610504, "learning_rate": 0.0002, "epoch": 0.4050355774493706, "step": 370}, {"loss": 1.1842, "grad_norm": 0.5123590230941772, "learning_rate": 0.0002, "epoch": 0.41598248494800216, "step": 380}, {"loss": 1.1539, "grad_norm": 0.3444574773311615, "learning_rate": 0.0002, "epoch": 0.4269293924466338, "step": 390}, {"loss": 1.1756, "grad_norm": 0.5302175283432007, "learning_rate": 0.0002, "epoch": 0.43787629994526545, "step": 400}, {"loss": 1.1138, "grad_norm": 0.2713572680950165, "learning_rate": 0.0002, "epoch": 0.4488232074438971, "step": 410}, {"loss": 1.1281, "grad_norm": 0.29530611634254456, "learning_rate": 0.0002, "epoch": 0.45977011494252873, "step": 420}, {"loss": 1.1721, "grad_norm": 0.27282455563545227, "learning_rate": 0.0002, "epoch": 0.47071702244116037, "step": 430}, {"loss": 1.1213, "grad_norm": 0.2647949755191803, "learning_rate": 0.0002, "epoch": 0.481663929939792, "step": 440}, {"loss": 1.1656, "grad_norm": 0.35509347915649414, "learning_rate": 0.0002, "epoch": 0.49261083743842365, "step": 450}, {"loss": 1.1251, "grad_norm": 0.1959609091281891, "learning_rate": 0.0002, "epoch": 0.5035577449370553, "step": 460}, {"loss": 1.1889, "grad_norm": 0.40090155601501465, "learning_rate": 0.0002, "epoch": 0.5145046524356869, "step": 470}, {"loss": 1.156, "grad_norm": 0.3354604244232178, "learning_rate": 0.0002, "epoch": 0.5254515599343186, "step": 480}, {"loss": 1.1024, "grad_norm": 0.2758506238460541, "learning_rate": 0.0002, "epoch": 0.5363984674329502, "step": 490}, {"loss": 1.1108, "grad_norm": 0.3629051744937897, "learning_rate": 0.0002, "epoch": 0.5473453749315819, "step": 500}, {"loss": 1.2236, "grad_norm": 0.30802229046821594, "learning_rate": 0.0002, "epoch": 0.5582922824302134, "step": 510}, {"loss": 1.0424, "grad_norm": 0.3099463880062103, "learning_rate": 0.0002, "epoch": 0.5692391899288451, "step": 520}, {"loss": 1.255, "grad_norm": 0.42299067974090576, "learning_rate": 0.0002, "epoch": 0.5801860974274767, "step": 530}, {"loss": 1.1698, "grad_norm": 0.5392252802848816, "learning_rate": 0.0002, "epoch": 0.5911330049261084, "step": 540}, {"loss": 1.1171, "grad_norm": 0.34768250584602356, "learning_rate": 0.0002, "epoch": 0.60207991242474, "step": 550}, {"loss": 1.102, "grad_norm": 0.28490015864372253, "learning_rate": 0.0002, "epoch": 0.6130268199233716, "step": 560}, {"loss": 1.1111, "grad_norm": 0.34787994623184204, "learning_rate": 0.0002, "epoch": 0.6239737274220033, "step": 570}, {"loss": 1.0759, "grad_norm": 0.29058772325515747, "learning_rate": 0.0002, "epoch": 0.6349206349206349, "step": 580}, {"loss": 1.1157, "grad_norm": 0.4063778817653656, "learning_rate": 0.0002, "epoch": 0.6458675424192666, "step": 590}, {"loss": 1.1432, "grad_norm": 0.9244267344474792, "learning_rate": 0.0002, "epoch": 0.6568144499178982, "step": 600}, {"loss": 1.0591, "grad_norm": 0.27605190873146057, "learning_rate": 0.0002, "epoch": 0.6677613574165299, "step": 610}, {"loss": 1.2123, "grad_norm": 0.34346821904182434, "learning_rate": 0.0002, "epoch": 0.6787082649151615, "step": 620}, {"loss": 1.2195, "grad_norm": 0.3093279302120209, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 630}, {"loss": 1.2461, "grad_norm": 1.0069009065628052, "learning_rate": 0.0002, "epoch": 0.7006020799124247, "step": 640}, {"loss": 1.0892, "grad_norm": 0.5049130916595459, "learning_rate": 0.0002, "epoch": 0.7115489874110563, "step": 650}, {"loss": 1.1404, "grad_norm": 0.3748924732208252, "learning_rate": 0.0002, "epoch": 0.722495894909688, "step": 660}, {"loss": 1.1062, "grad_norm": 0.2964959144592285, "learning_rate": 0.0002, "epoch": 0.7334428024083196, "step": 670}, {"loss": 1.2617, "grad_norm": 0.4599986970424652, "learning_rate": 0.0002, "epoch": 0.7443897099069513, "step": 680}, {"loss": 1.088, "grad_norm": 0.27292951941490173, "learning_rate": 0.0002, "epoch": 0.7553366174055829, "step": 690}, {"loss": 1.2047, "grad_norm": 0.3123566806316376, "learning_rate": 0.0002, "epoch": 0.7662835249042146, "step": 700}, {"loss": 1.0021, "grad_norm": 0.28310710191726685, "learning_rate": 0.0002, "epoch": 0.7772304324028462, "step": 710}, {"loss": 1.2281, "grad_norm": 0.3279992341995239, "learning_rate": 0.0002, "epoch": 0.7881773399014779, "step": 720}, {"loss": 1.25, "grad_norm": 0.28179168701171875, "learning_rate": 0.0002, "epoch": 0.7991242474001095, "step": 730}, {"loss": 1.0602, "grad_norm": 0.31492987275123596, "learning_rate": 0.0002, "epoch": 0.8100711548987412, "step": 740}, {"loss": 1.2518, "grad_norm": 0.41821011900901794, "learning_rate": 0.0002, "epoch": 0.8210180623973727, "step": 750}, {"loss": 1.1612, "grad_norm": 0.325235515832901, "learning_rate": 0.0002, "epoch": 0.8319649698960043, "step": 760}, {"loss": 1.27, "grad_norm": 0.5366070866584778, "learning_rate": 0.0002, "epoch": 0.842911877394636, "step": 770}, {"loss": 1.0921, "grad_norm": 0.32570579648017883, "learning_rate": 0.0002, "epoch": 0.8538587848932676, "step": 780}, {"loss": 1.1032, "grad_norm": 0.3642968237400055, "learning_rate": 0.0002, "epoch": 0.8648056923918993, "step": 790}, {"loss": 1.1234, "grad_norm": 0.29713448882102966, "learning_rate": 0.0002, "epoch": 0.8757525998905309, "step": 800}, {"loss": 1.0978, "grad_norm": 0.23599444329738617, "learning_rate": 0.0002, "epoch": 0.8866995073891626, "step": 810}, {"loss": 1.1867, "grad_norm": 0.31522464752197266, "learning_rate": 0.0002, "epoch": 0.8976464148877942, "step": 820}, {"loss": 1.0208, "grad_norm": 0.32754790782928467, "learning_rate": 0.0002, "epoch": 0.9085933223864259, "step": 830}, {"loss": 0.9786, "grad_norm": 0.22741089761257172, "learning_rate": 0.0002, "epoch": 0.9195402298850575, "step": 840}, {"loss": 1.0689, "grad_norm": 0.3089679777622223, "learning_rate": 0.0002, "epoch": 0.9304871373836892, "step": 850}, {"loss": 1.0354, "grad_norm": 0.27440521121025085, "learning_rate": 0.0002, "epoch": 0.9414340448823207, "step": 860}, {"loss": 1.0417, "grad_norm": 0.3498363792896271, "learning_rate": 0.0002, "epoch": 0.9523809523809523, "step": 870}, {"loss": 1.269, "grad_norm": 0.47151854634284973, "learning_rate": 0.0002, "epoch": 0.963327859879584, "step": 880}, {"loss": 1.1174, "grad_norm": 0.24756591022014618, "learning_rate": 0.0002, "epoch": 0.9742747673782156, "step": 890}, {"loss": 1.0622, "grad_norm": 0.2600938677787781, "learning_rate": 0.0002, "epoch": 0.9852216748768473, "step": 900}, {"loss": 1.0968, "grad_norm": 0.2934586703777313, "learning_rate": 0.0002, "epoch": 0.9961685823754789, "step": 910}, {"eval_loss": 1.158464789390564, "eval_runtime": 46.0774, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 0.9994526546250684, "step": 913}, {"loss": 1.0322, "grad_norm": 0.37776654958724976, "learning_rate": 0.0002, "epoch": 1.0071154898741106, "step": 920}, {"loss": 1.1346, "grad_norm": 0.31784629821777344, "learning_rate": 0.0002, "epoch": 1.0180623973727423, "step": 930}, {"loss": 0.8976, "grad_norm": 0.24244336783885956, "learning_rate": 0.0002, "epoch": 1.0290093048713738, "step": 940}, {"loss": 1.0172, "grad_norm": 0.3185454308986664, "learning_rate": 0.0002, "epoch": 1.0399562123700055, "step": 950}, {"loss": 1.1645, "grad_norm": 0.3589441478252411, "learning_rate": 0.0002, "epoch": 1.0509031198686372, "step": 960}, {"loss": 1.0317, "grad_norm": 0.38593578338623047, "learning_rate": 0.0002, "epoch": 1.0618500273672686, "step": 970}, {"loss": 0.9913, "grad_norm": 0.39694955945014954, "learning_rate": 0.0002, "epoch": 1.0727969348659003, "step": 980}, {"loss": 1.0893, "grad_norm": 0.469817191362381, "learning_rate": 0.0002, "epoch": 1.083743842364532, "step": 990}, {"loss": 0.983, "grad_norm": 0.2634755074977875, "learning_rate": 0.0002, "epoch": 1.0946907498631637, "step": 1000}, {"loss": 1.0144, "grad_norm": 0.43189436197280884, "learning_rate": 0.0002, "epoch": 1.1056376573617952, "step": 1010}, {"loss": 0.9663, "grad_norm": 0.5559977889060974, "learning_rate": 0.0002, "epoch": 1.116584564860427, "step": 1020}, {"loss": 1.0481, "grad_norm": 0.32100191712379456, "learning_rate": 0.0002, "epoch": 1.1275314723590586, "step": 1030}, {"loss": 1.1012, "grad_norm": 0.40179768204689026, "learning_rate": 0.0002, "epoch": 1.1384783798576903, "step": 1040}, {"loss": 1.0029, "grad_norm": 0.3659493029117584, "learning_rate": 0.0002, "epoch": 1.1494252873563218, "step": 1050}, {"loss": 0.9597, "grad_norm": 0.701704204082489, "learning_rate": 0.0002, "epoch": 1.1603721948549535, "step": 1060}, {"loss": 1.0204, "grad_norm": 0.3650563359260559, "learning_rate": 0.0002, "epoch": 1.1713191023535852, "step": 1070}, {"loss": 0.907, "grad_norm": 0.3191976249217987, "learning_rate": 0.0002, "epoch": 1.1822660098522166, "step": 1080}, {"loss": 1.0648, "grad_norm": 0.3615441918373108, "learning_rate": 0.0002, "epoch": 1.1932129173508483, "step": 1090}, {"loss": 1.0067, "grad_norm": 0.39474231004714966, "learning_rate": 0.0002, "epoch": 1.20415982484948, "step": 1100}, {"loss": 0.9852, "grad_norm": 0.3752822279930115, "learning_rate": 0.0002, "epoch": 1.2151067323481117, "step": 1110}, {"loss": 1.0067, "grad_norm": 0.4165991246700287, "learning_rate": 0.0002, "epoch": 1.2260536398467432, "step": 1120}, {"loss": 1.0244, "grad_norm": 0.5326506495475769, "learning_rate": 0.0002, "epoch": 1.237000547345375, "step": 1130}, {"loss": 1.0542, "grad_norm": 0.48845794796943665, "learning_rate": 0.0002, "epoch": 1.2479474548440066, "step": 1140}, {"loss": 1.0885, "grad_norm": 0.29910150170326233, "learning_rate": 0.0002, "epoch": 1.2588943623426383, "step": 1150}, {"loss": 1.2233, "grad_norm": 0.5069725513458252, "learning_rate": 0.0002, "epoch": 1.2698412698412698, "step": 1160}, {"loss": 1.0992, "grad_norm": 0.29500406980514526, "learning_rate": 0.0002, "epoch": 1.2807881773399015, "step": 1170}, {"loss": 1.0291, "grad_norm": 0.4711538851261139, "learning_rate": 0.0002, "epoch": 1.2917350848385332, "step": 1180}, {"loss": 1.0513, "grad_norm": 0.4203340709209442, "learning_rate": 0.0002, "epoch": 1.3026819923371646, "step": 1190}, {"loss": 1.1274, "grad_norm": 0.36101874709129333, "learning_rate": 0.0002, "epoch": 1.3136288998357963, "step": 1200}, {"loss": 1.085, "grad_norm": 0.4608800411224365, "learning_rate": 0.0002, "epoch": 1.324575807334428, "step": 1210}, {"loss": 1.0695, "grad_norm": 0.6570906639099121, "learning_rate": 0.0002, "epoch": 1.3355227148330597, "step": 1220}, {"loss": 0.9786, "grad_norm": 0.5352164506912231, "learning_rate": 0.0002, "epoch": 1.3464696223316914, "step": 1230}, {"loss": 1.0513, "grad_norm": 0.3885001242160797, "learning_rate": 0.0002, "epoch": 1.357416529830323, "step": 1240}, {"loss": 1.0611, "grad_norm": 0.2987913489341736, "learning_rate": 0.0002, "epoch": 1.3683634373289546, "step": 1250}, {"loss": 1.109, "grad_norm": 0.42070427536964417, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1260}, {"loss": 1.1092, "grad_norm": 0.5957782864570618, "learning_rate": 0.0002, "epoch": 1.3902572523262178, "step": 1270}, {"loss": 1.0554, "grad_norm": 0.32898882031440735, "learning_rate": 0.0002, "epoch": 1.4012041598248495, "step": 1280}, {"loss": 0.909, "grad_norm": 0.27624452114105225, "learning_rate": 0.0002, "epoch": 1.4121510673234812, "step": 1290}, {"loss": 1.024, "grad_norm": 0.49570828676223755, "learning_rate": 0.0002, "epoch": 1.4230979748221126, "step": 1300}, {"loss": 1.0471, "grad_norm": 0.26191383600234985, "learning_rate": 0.0002, "epoch": 1.4340448823207443, "step": 1310}, {"loss": 1.0788, "grad_norm": 0.35664042830467224, "learning_rate": 0.0002, "epoch": 1.444991789819376, "step": 1320}, {"loss": 1.0108, "grad_norm": 0.45126354694366455, "learning_rate": 0.0002, "epoch": 1.4559386973180077, "step": 1330}, {"loss": 0.9571, "grad_norm": 0.37318357825279236, "learning_rate": 0.0002, "epoch": 1.4668856048166394, "step": 1340}, {"loss": 1.0507, "grad_norm": 0.6428970098495483, "learning_rate": 0.0002, "epoch": 1.477832512315271, "step": 1350}, {"loss": 1.0367, "grad_norm": 0.43256187438964844, "learning_rate": 0.0002, "epoch": 1.4887794198139026, "step": 1360}, {"loss": 1.1321, "grad_norm": 0.5343793630599976, "learning_rate": 0.0002, "epoch": 1.4997263273125343, "step": 1370}, {"loss": 1.1054, "grad_norm": 0.315437376499176, "learning_rate": 0.0002, "epoch": 1.5106732348111658, "step": 1380}, {"loss": 0.8916, "grad_norm": 0.41561153531074524, "learning_rate": 0.0002, "epoch": 1.5216201423097975, "step": 1390}, {"loss": 1.0391, "grad_norm": 0.3201070725917816, "learning_rate": 0.0002, "epoch": 1.5325670498084292, "step": 1400}, {"loss": 1.14, "grad_norm": 0.505537211894989, "learning_rate": 0.0002, "epoch": 1.5435139573070606, "step": 1410}, {"loss": 1.0775, "grad_norm": 0.3747410178184509, "learning_rate": 0.0002, "epoch": 1.5544608648056923, "step": 1420}, {"loss": 1.1171, "grad_norm": 0.49385908246040344, "learning_rate": 0.0002, "epoch": 1.565407772304324, "step": 1430}, {"loss": 1.1182, "grad_norm": 0.49831628799438477, "learning_rate": 0.0002, "epoch": 1.5763546798029555, "step": 1440}, {"loss": 1.0079, "grad_norm": 0.372127890586853, "learning_rate": 0.0002, "epoch": 1.5873015873015874, "step": 1450}, {"loss": 0.9931, "grad_norm": 0.40070840716362, "learning_rate": 0.0002, "epoch": 1.598248494800219, "step": 1460}, {"loss": 0.8954, "grad_norm": 0.34907400608062744, "learning_rate": 0.0002, "epoch": 1.6091954022988506, "step": 1470}, {"loss": 0.9743, "grad_norm": 0.4632418751716614, "learning_rate": 0.0002, "epoch": 1.6201423097974823, "step": 1480}, {"loss": 1.0103, "grad_norm": 0.40164515376091003, "learning_rate": 0.0002, "epoch": 1.6310892172961138, "step": 1490}, {"loss": 0.9523, "grad_norm": 0.3214994966983795, "learning_rate": 0.0002, "epoch": 1.6420361247947455, "step": 1500}, {"loss": 1.0161, "grad_norm": 0.3727897107601166, "learning_rate": 0.0002, "epoch": 1.6529830322933772, "step": 1510}, {"loss": 1.0443, "grad_norm": 0.3817640542984009, "learning_rate": 0.0002, "epoch": 1.6639299397920086, "step": 1520}, {"loss": 1.0511, "grad_norm": 0.5592136979103088, "learning_rate": 0.0002, "epoch": 1.6748768472906403, "step": 1530}, {"loss": 0.9682, "grad_norm": 0.44636598229408264, "learning_rate": 0.0002, "epoch": 1.685823754789272, "step": 1540}, {"loss": 1.033, "grad_norm": 0.40441709756851196, "learning_rate": 0.0002, "epoch": 1.6967706622879035, "step": 1550}, {"loss": 1.0857, "grad_norm": 0.3243522644042969, "learning_rate": 0.0002, "epoch": 1.7077175697865354, "step": 1560}, {"loss": 0.9482, "grad_norm": 0.34277570247650146, "learning_rate": 0.0002, "epoch": 1.718664477285167, "step": 1570}, {"loss": 0.974, "grad_norm": 0.3279995024204254, "learning_rate": 0.0002, "epoch": 1.7296113847837986, "step": 1580}, {"loss": 0.9414, "grad_norm": 0.41968777775764465, "learning_rate": 0.0002, "epoch": 1.7405582922824303, "step": 1590}, {"loss": 0.9768, "grad_norm": 0.39464613795280457, "learning_rate": 0.0002, "epoch": 1.7515051997810618, "step": 1600}, {"loss": 1.0347, "grad_norm": 0.3839009404182434, "learning_rate": 0.0002, "epoch": 1.7624521072796935, "step": 1610}, {"loss": 0.9195, "grad_norm": 0.3250715434551239, "learning_rate": 0.0002, "epoch": 1.7733990147783252, "step": 1620}, {"loss": 1.0049, "grad_norm": 0.5166561007499695, "learning_rate": 0.0002, "epoch": 1.7843459222769567, "step": 1630}, {"loss": 1.0364, "grad_norm": 0.4115183353424072, "learning_rate": 0.0002, "epoch": 1.7952928297755884, "step": 1640}, {"loss": 1.0248, "grad_norm": 0.373780220746994, "learning_rate": 0.0002, "epoch": 1.80623973727422, "step": 1650}, {"loss": 1.0984, "grad_norm": 0.49697014689445496, "learning_rate": 0.0002, "epoch": 1.8171866447728515, "step": 1660}, {"loss": 1.0089, "grad_norm": 1.0308938026428223, "learning_rate": 0.0002, "epoch": 1.8281335522714834, "step": 1670}, {"loss": 1.0853, "grad_norm": 0.4851366877555847, "learning_rate": 0.0002, "epoch": 1.839080459770115, "step": 1680}, {"loss": 0.9533, "grad_norm": 0.3262481391429901, "learning_rate": 0.0002, "epoch": 1.8500273672687466, "step": 1690}, {"loss": 1.048, "grad_norm": 0.6904496550559998, "learning_rate": 0.0002, "epoch": 1.8609742747673783, "step": 1700}, {"loss": 1.0577, "grad_norm": 0.49789851903915405, "learning_rate": 0.0002, "epoch": 1.8719211822660098, "step": 1710}, {"loss": 1.0258, "grad_norm": 0.3035794198513031, "learning_rate": 0.0002, "epoch": 1.8828680897646415, "step": 1720}, {"loss": 0.9916, "grad_norm": 0.4588414430618286, "learning_rate": 0.0002, "epoch": 1.8938149972632732, "step": 1730}, {"loss": 0.9526, "grad_norm": 0.4313034117221832, "learning_rate": 0.0002, "epoch": 1.9047619047619047, "step": 1740}, {"loss": 1.0857, "grad_norm": 0.38562044501304626, "learning_rate": 0.0002, "epoch": 1.9157088122605364, "step": 1750}, {"loss": 0.9763, "grad_norm": 0.46947410702705383, "learning_rate": 0.0002, "epoch": 1.926655719759168, "step": 1760}, {"loss": 0.9967, "grad_norm": 0.3848404884338379, "learning_rate": 0.0002, "epoch": 1.9376026272577995, "step": 1770}, {"loss": 1.0474, "grad_norm": 0.30422744154930115, "learning_rate": 0.0002, "epoch": 1.9485495347564314, "step": 1780}, {"loss": 1.022, "grad_norm": 0.41100990772247314, "learning_rate": 0.0002, "epoch": 1.959496442255063, "step": 1790}, {"loss": 1.003, "grad_norm": 0.3492335081100464, "learning_rate": 0.0002, "epoch": 1.9704433497536946, "step": 1800}, {"loss": 0.9891, "grad_norm": 0.364577978849411, "learning_rate": 0.0002, "epoch": 1.9813902572523263, "step": 1810}, {"loss": 1.0699, "grad_norm": 0.4312075674533844, "learning_rate": 0.0002, "epoch": 1.9923371647509578, "step": 1820}, {"eval_loss": 1.14472496509552, "eval_runtime": 46.0786, "eval_samples_per_second": 9.462, "eval_steps_per_second": 1.194, "epoch": 2.0, "step": 1827}, {"loss": 1.0218, "grad_norm": 0.5989689230918884, "learning_rate": 0.0002, "epoch": 2.0032840722495897, "step": 1830}, {"loss": 0.9384, "grad_norm": 0.49720922112464905, "learning_rate": 0.0002, "epoch": 2.014230979748221, "step": 1840}, {"loss": 0.8482, "grad_norm": 0.42675456404685974, "learning_rate": 0.0002, "epoch": 2.0251778872468527, "step": 1850}, {"loss": 0.8487, "grad_norm": 0.4637208580970764, "learning_rate": 0.0002, "epoch": 2.0361247947454846, "step": 1860}, {"loss": 0.8212, "grad_norm": 0.8329976797103882, "learning_rate": 0.0002, "epoch": 2.047071702244116, "step": 1870}, {"loss": 0.9547, "grad_norm": 0.7869427800178528, "learning_rate": 0.0002, "epoch": 2.0580186097427475, "step": 1880}, {"loss": 0.9351, "grad_norm": 0.4927455186843872, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 1890}, {"loss": 0.8652, "grad_norm": 0.6264246702194214, "learning_rate": 0.0002, "epoch": 2.079912424740011, "step": 1900}, {"loss": 0.9095, "grad_norm": 1.1164122819900513, "learning_rate": 0.0002, "epoch": 2.0908593322386424, "step": 1910}, {"loss": 0.7823, "grad_norm": 0.5283981561660767, "learning_rate": 0.0002, "epoch": 2.1018062397372743, "step": 1920}, {"loss": 0.8065, "grad_norm": 0.45621731877326965, "learning_rate": 0.0002, "epoch": 2.112753147235906, "step": 1930}, {"loss": 0.9184, "grad_norm": 1.381791591644287, "learning_rate": 0.0002, "epoch": 2.1237000547345373, "step": 1940}, {"loss": 0.9006, "grad_norm": 0.5151259899139404, "learning_rate": 0.0002, "epoch": 2.134646962233169, "step": 1950}, {"loss": 0.8436, "grad_norm": 0.9806339740753174, "learning_rate": 0.0002, "epoch": 2.1455938697318007, "step": 1960}, {"loss": 0.8749, "grad_norm": 0.4734154939651489, "learning_rate": 0.0002, "epoch": 2.1565407772304326, "step": 1970}, {"loss": 0.9172, "grad_norm": 0.9553168416023254, "learning_rate": 0.0002, "epoch": 2.167487684729064, "step": 1980}, {"loss": 0.8047, "grad_norm": 0.5895838141441345, "learning_rate": 0.0002, "epoch": 2.1784345922276955, "step": 1990}, {"loss": 0.7841, "grad_norm": 0.4488855302333832, "learning_rate": 0.0002, "epoch": 2.1893814997263275, "step": 2000}, {"loss": 0.8205, "grad_norm": 1.0760235786437988, "learning_rate": 0.0002, "epoch": 2.200328407224959, "step": 2010}, {"loss": 0.7923, "grad_norm": 0.5038785338401794, "learning_rate": 0.0002, "epoch": 2.2112753147235904, "step": 2020}, {"loss": 0.8973, "grad_norm": 0.59819495677948, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 2030}, {"loss": 0.8009, "grad_norm": 0.5012075304985046, "learning_rate": 0.0002, "epoch": 2.233169129720854, "step": 2040}, {"loss": 0.9087, "grad_norm": 0.44978439807891846, "learning_rate": 0.0002, "epoch": 2.2441160372194853, "step": 2050}, {"loss": 0.823, "grad_norm": 0.5350462198257446, "learning_rate": 0.0002, "epoch": 2.255062944718117, "step": 2060}, {"loss": 0.8335, "grad_norm": 0.6020669937133789, "learning_rate": 0.0002, "epoch": 2.2660098522167487, "step": 2070}, {"loss": 0.8023, "grad_norm": 0.5246821045875549, "learning_rate": 0.0002, "epoch": 2.2769567597153806, "step": 2080}, {"loss": 0.8984, "grad_norm": 0.5711268782615662, "learning_rate": 0.0002, "epoch": 2.287903667214012, "step": 2090}, {"loss": 0.9093, "grad_norm": 0.617317259311676, "learning_rate": 0.0002, "epoch": 2.2988505747126435, "step": 2100}, {"loss": 0.8311, "grad_norm": 0.8608947396278381, "learning_rate": 0.0002, "epoch": 2.3097974822112755, "step": 2110}, {"loss": 0.7839, "grad_norm": 0.4739076793193817, "learning_rate": 0.0002, "epoch": 2.320744389709907, "step": 2120}, {"loss": 0.84, "grad_norm": 0.5538856983184814, "learning_rate": 0.0002, "epoch": 2.3316912972085384, "step": 2130}, {"loss": 0.8994, "grad_norm": 0.6064935326576233, "learning_rate": 0.0002, "epoch": 2.3426382047071703, "step": 2140}, {"loss": 0.7765, "grad_norm": 0.5019068121910095, "learning_rate": 0.0002, "epoch": 2.353585112205802, "step": 2150}, {"loss": 0.9576, "grad_norm": 0.45340514183044434, "learning_rate": 0.0002, "epoch": 2.3645320197044333, "step": 2160}, {"loss": 0.9254, "grad_norm": 0.7347203493118286, "learning_rate": 0.0002, "epoch": 2.375478927203065, "step": 2170}, {"loss": 0.8507, "grad_norm": 0.46922534704208374, "learning_rate": 0.0002, "epoch": 2.3864258347016967, "step": 2180}, {"loss": 0.8325, "grad_norm": 0.5507845878601074, "learning_rate": 0.0002, "epoch": 2.3973727422003286, "step": 2190}, {"loss": 0.8178, "grad_norm": 0.5621911883354187, "learning_rate": 0.0002, "epoch": 2.40831964969896, "step": 2200}, {"loss": 0.8569, "grad_norm": 0.5023514032363892, "learning_rate": 0.0002, "epoch": 2.4192665571975915, "step": 2210}, {"loss": 0.8355, "grad_norm": 0.6124861240386963, "learning_rate": 0.0002, "epoch": 2.4302134646962235, "step": 2220}, {"loss": 0.885, "grad_norm": 0.49614205956459045, "learning_rate": 0.0002, "epoch": 2.441160372194855, "step": 2230}, {"loss": 0.8008, "grad_norm": 0.6477900743484497, "learning_rate": 0.0002, "epoch": 2.4521072796934864, "step": 2240}, {"loss": 0.8622, "grad_norm": 0.5868843793869019, "learning_rate": 0.0002, "epoch": 2.4630541871921183, "step": 2250}, {"loss": 0.8498, "grad_norm": 0.4364610016345978, "learning_rate": 0.0002, "epoch": 2.47400109469075, "step": 2260}, {"loss": 0.8378, "grad_norm": 0.5792964696884155, "learning_rate": 0.0002, "epoch": 2.4849480021893813, "step": 2270}, {"loss": 0.8743, "grad_norm": 0.5421269536018372, "learning_rate": 0.0002, "epoch": 2.495894909688013, "step": 2280}, {"loss": 0.9637, "grad_norm": 0.5525493025779724, "learning_rate": 0.0002, "epoch": 2.5068418171866447, "step": 2290}, {"loss": 0.8075, "grad_norm": 0.6463850140571594, "learning_rate": 0.0002, "epoch": 2.5177887246852766, "step": 2300}, {"loss": 0.8591, "grad_norm": 0.6861311793327332, "learning_rate": 0.0002, "epoch": 2.528735632183908, "step": 2310}, {"loss": 0.9287, "grad_norm": 0.5563555359840393, "learning_rate": 0.0002, "epoch": 2.5396825396825395, "step": 2320}, {"loss": 0.945, "grad_norm": 0.5721169114112854, "learning_rate": 0.0002, "epoch": 2.5506294471811715, "step": 2330}, {"loss": 0.8271, "grad_norm": 0.5258274674415588, "learning_rate": 0.0002, "epoch": 2.561576354679803, "step": 2340}, {"loss": 0.8515, "grad_norm": 0.7057380676269531, "learning_rate": 0.0002, "epoch": 2.572523262178435, "step": 2350}, {"loss": 0.8615, "grad_norm": 0.6869027614593506, "learning_rate": 0.0002, "epoch": 2.5834701696770663, "step": 2360}, {"loss": 0.8043, "grad_norm": 0.4960809648036957, "learning_rate": 0.0002, "epoch": 2.594417077175698, "step": 2370}, {"loss": 0.8476, "grad_norm": 0.9288380146026611, "learning_rate": 0.0002, "epoch": 2.6053639846743293, "step": 2380}, {"loss": 0.873, "grad_norm": 0.3765334188938141, "learning_rate": 0.0002, "epoch": 2.616310892172961, "step": 2390}, {"loss": 0.8764, "grad_norm": 0.7487865686416626, "learning_rate": 0.0002, "epoch": 2.6272577996715927, "step": 2400}, {"loss": 0.7577, "grad_norm": 0.6141156554222107, "learning_rate": 0.0002, "epoch": 2.6382047071702246, "step": 2410}, {"loss": 0.8534, "grad_norm": 0.8420507907867432, "learning_rate": 0.0002, "epoch": 2.649151614668856, "step": 2420}, {"loss": 0.8311, "grad_norm": 0.53386390209198, "learning_rate": 0.0002, "epoch": 2.6600985221674875, "step": 2430}, {"loss": 0.8486, "grad_norm": 0.5520607233047485, "learning_rate": 0.0002, "epoch": 2.6710454296661195, "step": 2440}, {"loss": 0.8686, "grad_norm": 0.5337599515914917, "learning_rate": 0.0002, "epoch": 2.681992337164751, "step": 2450}, {"loss": 0.792, "grad_norm": 0.48790836334228516, "learning_rate": 0.0002, "epoch": 2.692939244663383, "step": 2460}, {"loss": 0.7864, "grad_norm": 0.8287786245346069, "learning_rate": 0.0002, "epoch": 2.7038861521620143, "step": 2470}, {"loss": 0.8244, "grad_norm": 0.5876168608665466, "learning_rate": 0.0002, "epoch": 2.714833059660646, "step": 2480}, {"loss": 0.8773, "grad_norm": 0.5206760764122009, "learning_rate": 0.0002, "epoch": 2.7257799671592773, "step": 2490}, {"loss": 0.8097, "grad_norm": 0.5619136691093445, "learning_rate": 0.0002, "epoch": 2.736726874657909, "step": 2500}, {"loss": 0.8377, "grad_norm": 0.5614883899688721, "learning_rate": 0.0002, "epoch": 2.7476737821565407, "step": 2510}, {"loss": 0.8817, "grad_norm": 0.6157700419425964, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 2520}, {"loss": 0.846, "grad_norm": 0.5529953837394714, "learning_rate": 0.0002, "epoch": 2.769567597153804, "step": 2530}, {"loss": 0.8881, "grad_norm": 0.6731224060058594, "learning_rate": 0.0002, "epoch": 2.7805145046524355, "step": 2540}, {"loss": 0.9125, "grad_norm": 0.6960386633872986, "learning_rate": 0.0002, "epoch": 2.7914614121510675, "step": 2550}, {"loss": 0.7823, "grad_norm": 0.5203493237495422, "learning_rate": 0.0002, "epoch": 2.802408319649699, "step": 2560}, {"loss": 0.7951, "grad_norm": 1.036837100982666, "learning_rate": 0.0002, "epoch": 2.813355227148331, "step": 2570}, {"loss": 0.8427, "grad_norm": 0.6125805377960205, "learning_rate": 0.0002, "epoch": 2.8243021346469623, "step": 2580}, {"loss": 0.8939, "grad_norm": 0.6298092603683472, "learning_rate": 0.0002, "epoch": 2.835249042145594, "step": 2590}, {"loss": 0.8845, "grad_norm": 0.5882203578948975, "learning_rate": 0.0002, "epoch": 2.8461959496442253, "step": 2600}, {"loss": 0.8921, "grad_norm": 0.8619399666786194, "learning_rate": 0.0002, "epoch": 2.857142857142857, "step": 2610}, {"loss": 0.8213, "grad_norm": 0.4722687304019928, "learning_rate": 0.0002, "epoch": 2.8680897646414887, "step": 2620}, {"loss": 0.9147, "grad_norm": 0.47399574518203735, "learning_rate": 0.0002, "epoch": 2.8790366721401206, "step": 2630}, {"loss": 0.8325, "grad_norm": 0.5639172792434692, "learning_rate": 0.0002, "epoch": 2.889983579638752, "step": 2640}, {"loss": 0.8653, "grad_norm": 0.4676816761493683, "learning_rate": 0.0002, "epoch": 2.9009304871373836, "step": 2650}, {"loss": 0.8966, "grad_norm": 0.6906291246414185, "learning_rate": 0.0002, "epoch": 2.9118773946360155, "step": 2660}, {"loss": 0.8966, "grad_norm": 0.4369746148586273, "learning_rate": 0.0002, "epoch": 2.922824302134647, "step": 2670}, {"loss": 0.9173, "grad_norm": 0.46423083543777466, "learning_rate": 0.0002, "epoch": 2.933771209633279, "step": 2680}, {"loss": 0.7961, "grad_norm": 0.5700525045394897, "learning_rate": 0.0002, "epoch": 2.9447181171319103, "step": 2690}, {"loss": 0.8172, "grad_norm": 0.6221476793289185, "learning_rate": 0.0002, "epoch": 2.955665024630542, "step": 2700}, {"loss": 0.8538, "grad_norm": 0.6102682948112488, "learning_rate": 0.0002, "epoch": 2.9666119321291733, "step": 2710}, {"loss": 0.9779, "grad_norm": 0.5317878723144531, "learning_rate": 0.0002, "epoch": 2.977558839627805, "step": 2720}, {"loss": 0.9314, "grad_norm": 0.4438510835170746, "learning_rate": 0.0002, "epoch": 2.9885057471264367, "step": 2730}, {"loss": 0.8553, "grad_norm": 0.5022130012512207, "learning_rate": 0.0002, "epoch": 2.9994526546250686, "step": 2740}, {"eval_loss": 1.1722838878631592, "eval_runtime": 46.0829, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.194, "epoch": 2.9994526546250686, "step": 2740}, {"loss": 0.6443, "grad_norm": 0.6384502053260803, "learning_rate": 0.0002, "epoch": 3.0103995621237, "step": 2750}, {"loss": 0.7123, "grad_norm": 0.9928722381591797, "learning_rate": 0.0002, "epoch": 3.0213464696223316, "step": 2760}, {"loss": 0.6045, "grad_norm": 0.7813051342964172, "learning_rate": 0.0002, "epoch": 3.0322933771209635, "step": 2770}, {"loss": 0.6042, "grad_norm": 1.0202556848526, "learning_rate": 0.0002, "epoch": 3.043240284619595, "step": 2780}, {"loss": 0.6356, "grad_norm": 0.7581062316894531, "learning_rate": 0.0002, "epoch": 3.0541871921182264, "step": 2790}, {"loss": 0.6349, "grad_norm": 0.6252710223197937, "learning_rate": 0.0002, "epoch": 3.0651340996168583, "step": 2800}, {"loss": 0.645, "grad_norm": 0.7738662958145142, "learning_rate": 0.0002, "epoch": 3.07608100711549, "step": 2810}, {"loss": 0.627, "grad_norm": 0.7381885051727295, "learning_rate": 0.0002, "epoch": 3.0870279146141213, "step": 2820}, {"loss": 0.6371, "grad_norm": 0.9197564721107483, "learning_rate": 0.0002, "epoch": 3.097974822112753, "step": 2830}, {"loss": 0.723, "grad_norm": 1.000976800918579, "learning_rate": 0.0002, "epoch": 3.1089217296113847, "step": 2840}, {"loss": 0.6631, "grad_norm": 0.7559131383895874, "learning_rate": 0.0002, "epoch": 3.1198686371100166, "step": 2850}, {"loss": 0.6252, "grad_norm": 0.7213780879974365, "learning_rate": 0.0002, "epoch": 3.130815544608648, "step": 2860}, {"loss": 0.6501, "grad_norm": 0.945939838886261, "learning_rate": 0.0002, "epoch": 3.1417624521072796, "step": 2870}, {"loss": 0.6129, "grad_norm": 0.7277454137802124, "learning_rate": 0.0002, "epoch": 3.1527093596059115, "step": 2880}, {"loss": 0.6423, "grad_norm": 0.762026846408844, "learning_rate": 0.0002, "epoch": 3.163656267104543, "step": 2890}, {"loss": 0.5332, "grad_norm": 0.6471221446990967, "learning_rate": 0.0002, "epoch": 3.1746031746031744, "step": 2900}, {"loss": 0.7981, "grad_norm": 0.6018978357315063, "learning_rate": 0.0002, "epoch": 3.1855500821018063, "step": 2910}, {"loss": 0.7274, "grad_norm": 0.8607320785522461, "learning_rate": 0.0002, "epoch": 3.196496989600438, "step": 2920}, {"loss": 0.6139, "grad_norm": 0.8854126334190369, "learning_rate": 0.0002, "epoch": 3.2074438970990693, "step": 2930}, {"loss": 0.6485, "grad_norm": 0.6620870232582092, "learning_rate": 0.0002, "epoch": 3.218390804597701, "step": 2940}, {"loss": 0.6969, "grad_norm": 0.7377511858940125, "learning_rate": 0.0002, "epoch": 3.2293377120963327, "step": 2950}, {"loss": 0.6798, "grad_norm": 0.7803301811218262, "learning_rate": 0.0002, "epoch": 3.2402846195949646, "step": 2960}, {"loss": 0.6697, "grad_norm": 0.834061861038208, "learning_rate": 0.0002, "epoch": 3.251231527093596, "step": 2970}, {"loss": 0.6894, "grad_norm": 0.8496041893959045, "learning_rate": 0.0002, "epoch": 3.2621784345922276, "step": 2980}, {"loss": 0.6591, "grad_norm": 0.7967984676361084, "learning_rate": 0.0002, "epoch": 3.2731253420908595, "step": 2990}, {"loss": 0.7266, "grad_norm": 1.0207016468048096, "learning_rate": 0.0002, "epoch": 3.284072249589491, "step": 3000}, {"loss": 0.6586, "grad_norm": 0.7855771780014038, "learning_rate": 0.0002, "epoch": 3.2950191570881224, "step": 3010}, {"loss": 0.5711, "grad_norm": 0.9427546858787537, "learning_rate": 0.0002, "epoch": 3.3059660645867543, "step": 3020}, {"loss": 0.6277, "grad_norm": 0.823542594909668, "learning_rate": 0.0002, "epoch": 3.316912972085386, "step": 3030}, {"loss": 0.7109, "grad_norm": 0.9826635122299194, "learning_rate": 0.0002, "epoch": 3.3278598795840173, "step": 3040}, {"loss": 0.6564, "grad_norm": 0.7259827852249146, "learning_rate": 0.0002, "epoch": 3.338806787082649, "step": 3050}, {"loss": 0.653, "grad_norm": 0.7774739861488342, "learning_rate": 0.0002, "epoch": 3.3497536945812807, "step": 3060}, {"loss": 0.7529, "grad_norm": 0.7394293546676636, "learning_rate": 0.0002, "epoch": 3.3607006020799126, "step": 3070}, {"loss": 0.5987, "grad_norm": 0.9017578959465027, "learning_rate": 0.0002, "epoch": 3.371647509578544, "step": 3080}, {"loss": 0.6953, "grad_norm": 0.7451054453849792, "learning_rate": 0.0002, "epoch": 3.3825944170771756, "step": 3090}, {"loss": 0.6759, "grad_norm": 0.7321506142616272, "learning_rate": 0.0002, "epoch": 3.3935413245758075, "step": 3100}, {"loss": 0.6555, "grad_norm": 0.6721828579902649, "learning_rate": 0.0002, "epoch": 3.404488232074439, "step": 3110}, {"loss": 0.6559, "grad_norm": 0.774022102355957, "learning_rate": 0.0002, "epoch": 3.4154351395730704, "step": 3120}, {"loss": 0.7449, "grad_norm": 0.9143537282943726, "learning_rate": 0.0002, "epoch": 3.4263820470717024, "step": 3130}, {"loss": 0.6899, "grad_norm": 1.226087212562561, "learning_rate": 0.0002, "epoch": 3.437328954570334, "step": 3140}, {"loss": 0.6719, "grad_norm": 0.7545496225357056, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 3150}, {"loss": 0.6153, "grad_norm": 0.6515635848045349, "learning_rate": 0.0002, "epoch": 3.4592227695675972, "step": 3160}, {"loss": 0.6926, "grad_norm": 0.9297090172767639, "learning_rate": 0.0002, "epoch": 3.4701696770662287, "step": 3170}, {"loss": 0.6071, "grad_norm": 1.0130730867385864, "learning_rate": 0.0002, "epoch": 3.4811165845648606, "step": 3180}, {"loss": 0.5959, "grad_norm": 0.7654589414596558, "learning_rate": 0.0002, "epoch": 3.492063492063492, "step": 3190}, {"loss": 0.7401, "grad_norm": 0.9954977631568909, "learning_rate": 0.0002, "epoch": 3.5030103995621236, "step": 3200}, {"loss": 0.6661, "grad_norm": 0.6027487516403198, "learning_rate": 0.0002, "epoch": 3.5139573070607555, "step": 3210}, {"loss": 0.6963, "grad_norm": 0.741770327091217, "learning_rate": 0.0002, "epoch": 3.524904214559387, "step": 3220}, {"loss": 0.8112, "grad_norm": 1.0534909963607788, "learning_rate": 0.0002, "epoch": 3.535851122058019, "step": 3230}, {"loss": 0.6813, "grad_norm": 0.937772274017334, "learning_rate": 0.0002, "epoch": 3.5467980295566504, "step": 3240}, {"loss": 0.6681, "grad_norm": 0.8504213690757751, "learning_rate": 0.0002, "epoch": 3.557744937055282, "step": 3250}, {"loss": 0.6436, "grad_norm": 0.7755007147789001, "learning_rate": 0.0002, "epoch": 3.5686918445539133, "step": 3260}, {"loss": 0.6213, "grad_norm": 1.0193358659744263, "learning_rate": 0.0002, "epoch": 3.5796387520525452, "step": 3270}, {"loss": 0.671, "grad_norm": 0.8440536856651306, "learning_rate": 0.0002, "epoch": 3.5905856595511767, "step": 3280}, {"loss": 0.6859, "grad_norm": 0.6195939183235168, "learning_rate": 0.0002, "epoch": 3.6015325670498086, "step": 3290}, {"loss": 0.7446, "grad_norm": 0.8608590960502625, "learning_rate": 0.0002, "epoch": 3.61247947454844, "step": 3300}, {"loss": 0.7301, "grad_norm": 0.6772327423095703, "learning_rate": 0.0002, "epoch": 3.6234263820470716, "step": 3310}, {"loss": 0.6298, "grad_norm": 0.8031839728355408, "learning_rate": 0.0002, "epoch": 3.6343732895457035, "step": 3320}, {"loss": 0.7041, "grad_norm": 0.6080502271652222, "learning_rate": 0.0002, "epoch": 3.645320197044335, "step": 3330}, {"loss": 0.7431, "grad_norm": 0.8007240891456604, "learning_rate": 0.0002, "epoch": 3.656267104542967, "step": 3340}, {"loss": 0.7446, "grad_norm": 0.8060704469680786, "learning_rate": 0.0002, "epoch": 3.6672140120415984, "step": 3350}, {"loss": 0.6304, "grad_norm": 0.7547586560249329, "learning_rate": 0.0002, "epoch": 3.67816091954023, "step": 3360}, {"loss": 0.7066, "grad_norm": 0.686851978302002, "learning_rate": 0.0002, "epoch": 3.6891078270388613, "step": 3370}, {"loss": 0.6748, "grad_norm": 0.9429075717926025, "learning_rate": 0.0002, "epoch": 3.7000547345374932, "step": 3380}, {"loss": 0.6673, "grad_norm": 0.7283591032028198, "learning_rate": 0.0002, "epoch": 3.7110016420361247, "step": 3390}, {"loss": 0.7502, "grad_norm": 0.8323085904121399, "learning_rate": 0.0002, "epoch": 3.7219485495347566, "step": 3400}, {"loss": 0.7779, "grad_norm": 0.8529590964317322, "learning_rate": 0.0002, "epoch": 3.732895457033388, "step": 3410}, {"loss": 0.6555, "grad_norm": 0.731752872467041, "learning_rate": 0.0002, "epoch": 3.7438423645320196, "step": 3420}, {"loss": 0.6928, "grad_norm": 0.8572278618812561, "learning_rate": 0.0002, "epoch": 3.7547892720306515, "step": 3430}, {"loss": 0.6215, "grad_norm": 0.7408691048622131, "learning_rate": 0.0002, "epoch": 3.765736179529283, "step": 3440}, {"loss": 0.622, "grad_norm": 0.7470445036888123, "learning_rate": 0.0002, "epoch": 3.776683087027915, "step": 3450}, {"loss": 0.7241, "grad_norm": 0.6806244254112244, "learning_rate": 0.0002, "epoch": 3.7876299945265464, "step": 3460}, {"loss": 0.7739, "grad_norm": 0.9129069447517395, "learning_rate": 0.0002, "epoch": 3.798576902025178, "step": 3470}, {"loss": 0.6826, "grad_norm": 0.8717501759529114, "learning_rate": 0.0002, "epoch": 3.8095238095238093, "step": 3480}, {"loss": 0.6188, "grad_norm": 0.6761979460716248, "learning_rate": 0.0002, "epoch": 3.8204707170224412, "step": 3490}, {"loss": 0.7601, "grad_norm": 1.0054380893707275, "learning_rate": 0.0002, "epoch": 3.8314176245210727, "step": 3500}, {"loss": 0.623, "grad_norm": 1.1224009990692139, "learning_rate": 0.0002, "epoch": 3.8423645320197046, "step": 3510}, {"loss": 0.6918, "grad_norm": 0.8997692465782166, "learning_rate": 0.0002, "epoch": 3.853311439518336, "step": 3520}, {"loss": 0.6357, "grad_norm": 1.0086902379989624, "learning_rate": 0.0002, "epoch": 3.8642583470169676, "step": 3530}, {"loss": 0.6379, "grad_norm": 0.772739589214325, "learning_rate": 0.0002, "epoch": 3.8752052545155995, "step": 3540}, {"loss": 0.7423, "grad_norm": 1.211774230003357, "learning_rate": 0.0002, "epoch": 3.886152162014231, "step": 3550}, {"loss": 0.7321, "grad_norm": 0.9572356939315796, "learning_rate": 0.0002, "epoch": 3.897099069512863, "step": 3560}, {"loss": 0.6836, "grad_norm": 0.7887842655181885, "learning_rate": 0.0002, "epoch": 3.9080459770114944, "step": 3570}, {"loss": 0.7576, "grad_norm": 0.7308389544487, "learning_rate": 0.0002, "epoch": 3.918992884510126, "step": 3580}, {"loss": 0.6001, "grad_norm": 1.0182650089263916, "learning_rate": 0.0002, "epoch": 3.9299397920087573, "step": 3590}, {"loss": 0.6942, "grad_norm": 0.8000147342681885, "learning_rate": 0.0002, "epoch": 3.9408866995073892, "step": 3600}, {"loss": 0.6244, "grad_norm": 0.7385728359222412, "learning_rate": 0.0002, "epoch": 3.9518336070060207, "step": 3610}, {"loss": 0.6718, "grad_norm": 0.9233261942863464, "learning_rate": 0.0002, "epoch": 3.9627805145046526, "step": 3620}, {"loss": 0.6508, "grad_norm": 0.8486751914024353, "learning_rate": 0.0002, "epoch": 3.973727422003284, "step": 3630}, {"loss": 0.6928, "grad_norm": 0.7593663334846497, "learning_rate": 0.0002, "epoch": 3.9846743295019156, "step": 3640}, {"loss": 0.6847, "grad_norm": 0.7885415554046631, "learning_rate": 0.0002, "epoch": 3.9956212370005475, "step": 3650}, {"eval_loss": 1.250312328338623, "eval_runtime": 46.0842, "eval_samples_per_second": 9.461, "eval_steps_per_second": 1.193, "epoch": 4.0, "step": 3654}, {"loss": 0.5547, "grad_norm": 0.6591703295707703, "learning_rate": 0.0002, "epoch": 4.006568144499179, "step": 3660}, {"loss": 0.5301, "grad_norm": 1.36927330493927, "learning_rate": 0.0002, "epoch": 4.017515051997811, "step": 3670}, {"loss": 0.4466, "grad_norm": 0.8106328845024109, "learning_rate": 0.0002, "epoch": 4.028461959496442, "step": 3680}, {"loss": 0.4861, "grad_norm": 0.7592712044715881, "learning_rate": 0.0002, "epoch": 4.039408866995074, "step": 3690}, {"loss": 0.5103, "grad_norm": 0.9518909454345703, "learning_rate": 0.0002, "epoch": 4.050355774493705, "step": 3700}, {"loss": 0.4638, "grad_norm": 0.7805967330932617, "learning_rate": 0.0002, "epoch": 4.061302681992337, "step": 3710}, {"loss": 0.4556, "grad_norm": 1.3146334886550903, "learning_rate": 0.0002, "epoch": 4.072249589490969, "step": 3720}, {"loss": 0.5635, "grad_norm": 1.1611138582229614, "learning_rate": 0.0002, "epoch": 4.083196496989601, "step": 3730}, {"loss": 0.3845, "grad_norm": 0.8173232078552246, "learning_rate": 0.0002, "epoch": 4.094143404488232, "step": 3740}, {"loss": 0.4911, "grad_norm": 0.7848323583602905, "learning_rate": 0.0002, "epoch": 4.105090311986864, "step": 3750}, {"loss": 0.4519, "grad_norm": 1.3183201551437378, "learning_rate": 0.0002, "epoch": 4.116037219485495, "step": 3760}, {"loss": 0.5083, "grad_norm": 1.1936529874801636, "learning_rate": 0.0002, "epoch": 4.1269841269841265, "step": 3770}, {"loss": 0.5208, "grad_norm": 1.1078993082046509, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 3780}, {"loss": 0.5928, "grad_norm": 1.107743263244629, "learning_rate": 0.0002, "epoch": 4.14887794198139, "step": 3790}, {"loss": 0.5112, "grad_norm": 0.7801875472068787, "learning_rate": 0.0002, "epoch": 4.159824849480022, "step": 3800}, {"loss": 0.4896, "grad_norm": 1.1328117847442627, "learning_rate": 0.0002, "epoch": 4.170771756978653, "step": 3810}, {"loss": 0.5645, "grad_norm": 1.4232193231582642, "learning_rate": 0.0002, "epoch": 4.181718664477285, "step": 3820}, {"loss": 0.5049, "grad_norm": 1.557416558265686, "learning_rate": 0.0002, "epoch": 4.192665571975917, "step": 3830}, {"loss": 0.4863, "grad_norm": 1.042923092842102, "learning_rate": 0.0002, "epoch": 4.203612479474549, "step": 3840}, {"loss": 0.3751, "grad_norm": 1.1801949739456177, "learning_rate": 0.0002, "epoch": 4.21455938697318, "step": 3850}, {"loss": 0.5063, "grad_norm": 0.9273753762245178, "learning_rate": 0.0002, "epoch": 4.225506294471812, "step": 3860}, {"loss": 0.5542, "grad_norm": 0.7681763768196106, "learning_rate": 0.0002, "epoch": 4.236453201970443, "step": 3870}, {"loss": 0.5971, "grad_norm": 0.9840841293334961, "learning_rate": 0.0002, "epoch": 4.2474001094690745, "step": 3880}, {"loss": 0.4648, "grad_norm": 1.0290725231170654, "learning_rate": 0.0002, "epoch": 4.258347016967707, "step": 3890}, {"loss": 0.4288, "grad_norm": 0.8059597611427307, "learning_rate": 0.0002, "epoch": 4.269293924466338, "step": 3900}, {"loss": 0.5103, "grad_norm": 0.9847467541694641, "learning_rate": 0.0002, "epoch": 4.28024083196497, "step": 3910}, {"loss": 0.4952, "grad_norm": 1.344044804573059, "learning_rate": 0.0002, "epoch": 4.291187739463601, "step": 3920}, {"loss": 0.4966, "grad_norm": 0.9174224138259888, "learning_rate": 0.0002, "epoch": 4.302134646962233, "step": 3930}, {"loss": 0.4944, "grad_norm": 1.1199711561203003, "learning_rate": 0.0002, "epoch": 4.313081554460865, "step": 3940}, {"loss": 0.4641, "grad_norm": 1.0120296478271484, "learning_rate": 0.0002, "epoch": 4.324028461959497, "step": 3950}, {"loss": 0.4723, "grad_norm": 1.091811180114746, "learning_rate": 0.0002, "epoch": 4.334975369458128, "step": 3960}, {"loss": 0.4627, "grad_norm": 1.0332133769989014, "learning_rate": 0.0002, "epoch": 4.34592227695676, "step": 3970}, {"loss": 0.4646, "grad_norm": 1.0785295963287354, "learning_rate": 0.0002, "epoch": 4.356869184455391, "step": 3980}, {"loss": 0.4909, "grad_norm": 1.0506969690322876, "learning_rate": 0.0002, "epoch": 4.3678160919540225, "step": 3990}, {"loss": 0.4776, "grad_norm": 1.047560691833496, "learning_rate": 0.0002, "epoch": 4.378762999452655, "step": 4000}, {"loss": 0.4549, "grad_norm": 0.9348800778388977, "learning_rate": 0.0002, "epoch": 4.389709906951286, "step": 4010}, {"loss": 0.5333, "grad_norm": 1.1563059091567993, "learning_rate": 0.0002, "epoch": 4.400656814449918, "step": 4020}, {"loss": 0.4952, "grad_norm": 1.001470923423767, "learning_rate": 0.0002, "epoch": 4.411603721948549, "step": 4030}, {"loss": 0.4972, "grad_norm": 1.309012532234192, "learning_rate": 0.0002, "epoch": 4.422550629447181, "step": 4040}, {"loss": 0.5078, "grad_norm": 0.7338925004005432, "learning_rate": 0.0002, "epoch": 4.433497536945813, "step": 4050}, {"loss": 0.4632, "grad_norm": 1.0398834943771362, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 4060}, {"loss": 0.6285, "grad_norm": 0.9728689193725586, "learning_rate": 0.0002, "epoch": 4.455391351943076, "step": 4070}, {"loss": 0.4778, "grad_norm": 1.247475028038025, "learning_rate": 0.0002, "epoch": 4.466338259441708, "step": 4080}, {"loss": 0.4813, "grad_norm": 1.1084578037261963, "learning_rate": 0.0002, "epoch": 4.477285166940339, "step": 4090}, {"loss": 0.5665, "grad_norm": 1.1619318723678589, "learning_rate": 0.0002, "epoch": 4.4882320744389705, "step": 4100}, {"loss": 0.5207, "grad_norm": 1.3456498384475708, "learning_rate": 0.0002, "epoch": 4.499178981937603, "step": 4110}, {"loss": 0.4876, "grad_norm": 0.9372991323471069, "learning_rate": 0.0002, "epoch": 4.510125889436234, "step": 4120}, {"loss": 0.5456, "grad_norm": 1.0071815252304077, "learning_rate": 0.0002, "epoch": 4.521072796934866, "step": 4130}, {"loss": 0.5589, "grad_norm": 1.190344214439392, "learning_rate": 0.0002, "epoch": 4.532019704433497, "step": 4140}, {"loss": 0.4852, "grad_norm": 0.9480887055397034, "learning_rate": 0.0002, "epoch": 4.542966611932129, "step": 4150}, {"loss": 0.5229, "grad_norm": 1.0252189636230469, "learning_rate": 0.0002, "epoch": 4.553913519430761, "step": 4160}, {"loss": 0.5253, "grad_norm": 0.7142013311386108, "learning_rate": 0.0002, "epoch": 4.564860426929393, "step": 4170}, {"loss": 0.4861, "grad_norm": 0.8937426805496216, "learning_rate": 0.0002, "epoch": 4.575807334428024, "step": 4180}, {"loss": 0.4773, "grad_norm": 0.8885005116462708, "learning_rate": 0.0002, "epoch": 4.586754241926656, "step": 4190}, {"loss": 0.4858, "grad_norm": 1.337663173675537, "learning_rate": 0.0002, "epoch": 4.597701149425287, "step": 4200}, {"loss": 0.5247, "grad_norm": 1.0475375652313232, "learning_rate": 0.0002, "epoch": 4.6086480569239185, "step": 4210}, {"loss": 0.5298, "grad_norm": 1.0081088542938232, "learning_rate": 0.0002, "epoch": 4.619594964422551, "step": 4220}, {"loss": 0.5042, "grad_norm": 0.7527595162391663, "learning_rate": 0.0002, "epoch": 4.630541871921182, "step": 4230}, {"loss": 0.5207, "grad_norm": 1.55559241771698, "learning_rate": 0.0002, "epoch": 4.641488779419814, "step": 4240}, {"loss": 0.5468, "grad_norm": 0.7967379689216614, "learning_rate": 0.0002, "epoch": 4.652435686918445, "step": 4250}, {"loss": 0.5328, "grad_norm": 0.898368775844574, "learning_rate": 0.0002, "epoch": 4.663382594417077, "step": 4260}, {"loss": 0.4706, "grad_norm": 1.1940776109695435, "learning_rate": 0.0002, "epoch": 4.674329501915709, "step": 4270}, {"loss": 0.5121, "grad_norm": 1.1817092895507812, "learning_rate": 0.0002, "epoch": 4.685276409414341, "step": 4280}, {"loss": 0.5758, "grad_norm": 0.9041520357131958, "learning_rate": 0.0002, "epoch": 4.696223316912972, "step": 4290}, {"loss": 0.5851, "grad_norm": 1.1280102729797363, "learning_rate": 0.0002, "epoch": 4.707170224411604, "step": 4300}, {"loss": 0.4891, "grad_norm": 1.357689619064331, "learning_rate": 0.0002, "epoch": 4.718117131910235, "step": 4310}, {"loss": 0.4704, "grad_norm": 1.056633472442627, "learning_rate": 0.0002, "epoch": 4.7290640394088665, "step": 4320}, {"loss": 0.5488, "grad_norm": 1.6520427465438843, "learning_rate": 0.0002, "epoch": 4.740010946907499, "step": 4330}, {"loss": 0.5131, "grad_norm": 1.153200626373291, "learning_rate": 0.0002, "epoch": 4.75095785440613, "step": 4340}, {"loss": 0.539, "grad_norm": 0.9346241354942322, "learning_rate": 0.0002, "epoch": 4.761904761904762, "step": 4350}, {"loss": 0.4941, "grad_norm": 0.8628455996513367, "learning_rate": 0.0002, "epoch": 4.772851669403393, "step": 4360}, {"loss": 0.5167, "grad_norm": 1.3843916654586792, "learning_rate": 0.0002, "epoch": 4.783798576902025, "step": 4370}, {"loss": 0.4683, "grad_norm": 1.035574197769165, "learning_rate": 0.0002, "epoch": 4.794745484400657, "step": 4380}, {"loss": 0.5162, "grad_norm": 1.1868361234664917, "learning_rate": 0.0002, "epoch": 4.805692391899289, "step": 4390}, {"loss": 0.534, "grad_norm": 1.1307647228240967, "learning_rate": 0.0002, "epoch": 4.81663929939792, "step": 4400}, {"loss": 0.5567, "grad_norm": 0.9787724614143372, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 4410}, {"loss": 0.5185, "grad_norm": 1.0473824739456177, "learning_rate": 0.0002, "epoch": 4.838533114395183, "step": 4420}, {"loss": 0.6285, "grad_norm": 1.069069504737854, "learning_rate": 0.0002, "epoch": 4.8494800218938146, "step": 4430}, {"loss": 0.5267, "grad_norm": 1.4305680990219116, "learning_rate": 0.0002, "epoch": 4.860426929392447, "step": 4440}, {"loss": 0.5947, "grad_norm": 1.3679203987121582, "learning_rate": 0.0002, "epoch": 4.871373836891078, "step": 4450}, {"loss": 0.5135, "grad_norm": 0.8997844457626343, "learning_rate": 0.0002, "epoch": 4.88232074438971, "step": 4460}, {"loss": 0.5312, "grad_norm": 1.2758110761642456, "learning_rate": 0.0002, "epoch": 4.893267651888341, "step": 4470}, {"loss": 0.4914, "grad_norm": 0.8819465637207031, "learning_rate": 0.0002, "epoch": 4.904214559386973, "step": 4480}, {"loss": 0.5147, "grad_norm": 1.08329439163208, "learning_rate": 0.0002, "epoch": 4.915161466885605, "step": 4490}, {"loss": 0.5404, "grad_norm": 1.083461046218872, "learning_rate": 0.0002, "epoch": 4.926108374384237, "step": 4500}, {"loss": 0.5433, "grad_norm": 1.2387723922729492, "learning_rate": 0.0002, "epoch": 4.937055281882868, "step": 4510}, {"loss": 0.5624, "grad_norm": 0.8262293934822083, "learning_rate": 0.0002, "epoch": 4.9480021893815, "step": 4520}, {"loss": 0.504, "grad_norm": 1.2325191497802734, "learning_rate": 0.0002, "epoch": 4.958949096880131, "step": 4530}, {"loss": 0.5452, "grad_norm": 1.024614930152893, "learning_rate": 0.0002, "epoch": 4.9698960043787626, "step": 4540}, {"loss": 0.4752, "grad_norm": 1.3007521629333496, "learning_rate": 0.0002, "epoch": 4.980842911877395, "step": 4550}, {"loss": 0.4943, "grad_norm": 0.9823828339576721, "learning_rate": 0.0002, "epoch": 4.991789819376026, "step": 4560}, {"eval_loss": 1.3920727968215942, "eval_runtime": 46.0764, "eval_samples_per_second": 9.463, "eval_steps_per_second": 1.194, "epoch": 4.999452654625069, "step": 4567}, {"loss": 0.545, "grad_norm": 1.1478906869888306, "learning_rate": 0.0002, "epoch": 5.002736726874658, "step": 4570}, {"loss": 0.372, "grad_norm": 1.0533705949783325, "learning_rate": 0.0002, "epoch": 5.013683634373289, "step": 4580}, {"loss": 0.3313, "grad_norm": 1.268900752067566, "learning_rate": 0.0002, "epoch": 5.024630541871921, "step": 4590}, {"loss": 0.3482, "grad_norm": 1.222652554512024, "learning_rate": 0.0002, "epoch": 5.035577449370553, "step": 4600}, {"loss": 0.3195, "grad_norm": 1.5093127489089966, "learning_rate": 0.0002, "epoch": 5.046524356869185, "step": 4610}, {"loss": 0.3569, "grad_norm": 1.2372499704360962, "learning_rate": 0.0002, "epoch": 5.057471264367816, "step": 4620}, {"loss": 0.3206, "grad_norm": 0.8422666192054749, "learning_rate": 0.0002, "epoch": 5.068418171866448, "step": 4630}, {"loss": 0.3115, "grad_norm": 1.1451770067214966, "learning_rate": 0.0002, "epoch": 5.079365079365079, "step": 4640}, {"loss": 0.3305, "grad_norm": 1.2074557542800903, "learning_rate": 0.0002, "epoch": 5.090311986863711, "step": 4650}, {"loss": 0.3012, "grad_norm": 1.429150104522705, "learning_rate": 0.0002, "epoch": 5.101258894362343, "step": 4660}, {"loss": 0.3229, "grad_norm": 1.0353610515594482, "learning_rate": 0.0002, "epoch": 5.112205801860974, "step": 4670}, {"loss": 0.402, "grad_norm": 1.2845979928970337, "learning_rate": 0.0002, "epoch": 5.123152709359606, "step": 4680}, {"loss": 0.383, "grad_norm": 1.3790186643600464, "learning_rate": 0.0002, "epoch": 5.134099616858237, "step": 4690}, {"loss": 0.2951, "grad_norm": 1.3182239532470703, "learning_rate": 0.0002, "epoch": 5.145046524356869, "step": 4700}, {"loss": 0.4074, "grad_norm": 1.5249626636505127, "learning_rate": 0.0002, "epoch": 5.155993431855501, "step": 4710}, {"loss": 0.3703, "grad_norm": 1.2492733001708984, "learning_rate": 0.0002, "epoch": 5.166940339354133, "step": 4720}, {"loss": 0.3411, "grad_norm": 1.4455480575561523, "learning_rate": 0.0002, "epoch": 5.177887246852764, "step": 4730}, {"loss": 0.3996, "grad_norm": 1.2191482782363892, "learning_rate": 0.0002, "epoch": 5.188834154351396, "step": 4740}, {"loss": 0.3785, "grad_norm": 1.4707951545715332, "learning_rate": 0.0002, "epoch": 5.199781061850027, "step": 4750}, {"loss": 0.3516, "grad_norm": 1.3473678827285767, "learning_rate": 0.0002, "epoch": 5.210727969348659, "step": 4760}, {"loss": 0.3266, "grad_norm": 1.0479670763015747, "learning_rate": 0.0002, "epoch": 5.221674876847291, "step": 4770}, {"loss": 0.3976, "grad_norm": 1.299096703529358, "learning_rate": 0.0002, "epoch": 5.232621784345922, "step": 4780}, {"loss": 0.3266, "grad_norm": 1.2820168733596802, "learning_rate": 0.0002, "epoch": 5.243568691844554, "step": 4790}, {"loss": 0.3347, "grad_norm": 1.3818004131317139, "learning_rate": 0.0002, "epoch": 5.254515599343185, "step": 4800}, {"loss": 0.3761, "grad_norm": 1.2898736000061035, "learning_rate": 0.0002, "epoch": 5.265462506841817, "step": 4810}, {"loss": 0.3694, "grad_norm": 1.1761468648910522, "learning_rate": 0.0002, "epoch": 5.276409414340449, "step": 4820}, {"loss": 0.3806, "grad_norm": 1.7155952453613281, "learning_rate": 0.0002, "epoch": 5.287356321839081, "step": 4830}, {"loss": 0.322, "grad_norm": 0.9103642106056213, "learning_rate": 0.0002, "epoch": 5.298303229337712, "step": 4840}, {"loss": 0.3516, "grad_norm": 1.013015627861023, "learning_rate": 0.0002, "epoch": 5.309250136836344, "step": 4850}, {"loss": 0.4297, "grad_norm": 1.390471339225769, "learning_rate": 0.0002, "epoch": 5.320197044334975, "step": 4860}, {"loss": 0.4098, "grad_norm": 1.129770278930664, "learning_rate": 0.0002, "epoch": 5.331143951833607, "step": 4870}, {"loss": 0.4227, "grad_norm": 1.1461067199707031, "learning_rate": 0.0002, "epoch": 5.342090859332239, "step": 4880}, {"loss": 0.288, "grad_norm": 1.3587424755096436, "learning_rate": 0.0002, "epoch": 5.35303776683087, "step": 4890}, {"loss": 0.3604, "grad_norm": 1.6897879838943481, "learning_rate": 0.0002, "epoch": 5.363984674329502, "step": 4900}, {"loss": 0.3887, "grad_norm": 0.9298055768013, "learning_rate": 0.0002, "epoch": 5.374931581828133, "step": 4910}, {"loss": 0.3371, "grad_norm": 1.0006917715072632, "learning_rate": 0.0002, "epoch": 5.385878489326765, "step": 4920}, {"loss": 0.3992, "grad_norm": 1.232581377029419, "learning_rate": 0.0002, "epoch": 5.396825396825397, "step": 4930}, {"loss": 0.3456, "grad_norm": 1.0822620391845703, "learning_rate": 0.0002, "epoch": 5.407772304324029, "step": 4940}, {"loss": 0.3806, "grad_norm": 1.3648720979690552, "learning_rate": 0.0002, "epoch": 5.41871921182266, "step": 4950}, {"loss": 0.3959, "grad_norm": 1.3220354318618774, "learning_rate": 0.0002, "epoch": 5.429666119321292, "step": 4960}, {"loss": 0.3278, "grad_norm": 1.1106271743774414, "learning_rate": 0.0002, "epoch": 5.440613026819923, "step": 4970}, {"loss": 0.3812, "grad_norm": 1.6058908700942993, "learning_rate": 0.0002, "epoch": 5.451559934318555, "step": 4980}, {"loss": 0.3905, "grad_norm": 1.1065930128097534, "learning_rate": 0.0002, "epoch": 5.462506841817187, "step": 4990}, {"loss": 0.4058, "grad_norm": 1.3896466493606567, "learning_rate": 0.0002, "epoch": 5.473453749315818, "step": 5000}, {"loss": 0.4122, "grad_norm": 1.0437148809432983, "learning_rate": 0.0002, "epoch": 5.48440065681445, "step": 5010}, {"loss": 0.4065, "grad_norm": 1.2347718477249146, "learning_rate": 0.0002, "epoch": 5.495347564313081, "step": 5020}, {"loss": 0.3586, "grad_norm": 1.1174284219741821, "learning_rate": 0.0002, "epoch": 5.506294471811713, "step": 5030}, {"loss": 0.3576, "grad_norm": 1.2580941915512085, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 5040}, {"loss": 0.3809, "grad_norm": 1.451090931892395, "learning_rate": 0.0002, "epoch": 5.528188286808977, "step": 5050}, {"loss": 0.3645, "grad_norm": 1.4688365459442139, "learning_rate": 0.0002, "epoch": 5.539135194307608, "step": 5060}, {"loss": 0.4431, "grad_norm": 1.1625734567642212, "learning_rate": 0.0002, "epoch": 5.55008210180624, "step": 5070}, {"loss": 0.3972, "grad_norm": 0.9332265257835388, "learning_rate": 0.0002, "epoch": 5.561029009304871, "step": 5080}, {"loss": 0.4, "grad_norm": 1.5635273456573486, "learning_rate": 0.0002, "epoch": 5.571975916803503, "step": 5090}, {"loss": 0.3651, "grad_norm": 1.3420509099960327, "learning_rate": 0.0002, "epoch": 5.582922824302135, "step": 5100}, {"loss": 0.3717, "grad_norm": 1.5826557874679565, "learning_rate": 0.0002, "epoch": 5.593869731800766, "step": 5110}, {"loss": 0.4256, "grad_norm": 1.5737065076828003, "learning_rate": 0.0002, "epoch": 5.604816639299398, "step": 5120}, {"loss": 0.39, "grad_norm": 1.3812499046325684, "learning_rate": 0.0002, "epoch": 5.615763546798029, "step": 5130}, {"loss": 0.3891, "grad_norm": 1.362833023071289, "learning_rate": 0.0002, "epoch": 5.626710454296661, "step": 5140}, {"loss": 0.455, "grad_norm": 1.7667874097824097, "learning_rate": 0.0002, "epoch": 5.637657361795293, "step": 5150}, {"loss": 0.4264, "grad_norm": 1.2661789655685425, "learning_rate": 0.0002, "epoch": 5.648604269293925, "step": 5160}, {"loss": 0.3261, "grad_norm": 1.2076870203018188, "learning_rate": 0.0002, "epoch": 5.659551176792556, "step": 5170}, {"loss": 0.372, "grad_norm": 1.2431524991989136, "learning_rate": 0.0002, "epoch": 5.670498084291188, "step": 5180}, {"loss": 0.4092, "grad_norm": 1.2216639518737793, "learning_rate": 0.0002, "epoch": 5.681444991789819, "step": 5190}, {"loss": 0.4171, "grad_norm": 0.9259352684020996, "learning_rate": 0.0002, "epoch": 5.692391899288451, "step": 5200}, {"loss": 0.3875, "grad_norm": 1.7929338216781616, "learning_rate": 0.0002, "epoch": 5.703338806787083, "step": 5210}, {"loss": 0.4424, "grad_norm": 1.4048460721969604, "learning_rate": 0.0002, "epoch": 5.714285714285714, "step": 5220}, {"loss": 0.3758, "grad_norm": 1.306874394416809, "learning_rate": 0.0002, "epoch": 5.725232621784346, "step": 5230}, {"loss": 0.3889, "grad_norm": 1.3137940168380737, "learning_rate": 0.0002, "epoch": 5.736179529282977, "step": 5240}, {"loss": 0.4804, "grad_norm": 1.1376476287841797, "learning_rate": 0.0002, "epoch": 5.747126436781609, "step": 5250}, {"loss": 0.377, "grad_norm": 1.450939416885376, "learning_rate": 0.0002, "epoch": 5.758073344280241, "step": 5260}, {"loss": 0.4732, "grad_norm": 0.983195960521698, "learning_rate": 0.0002, "epoch": 5.769020251778873, "step": 5270}, {"loss": 0.4041, "grad_norm": 1.66558837890625, "learning_rate": 0.0002, "epoch": 5.779967159277504, "step": 5280}, {"loss": 0.3643, "grad_norm": 0.9789204597473145, "learning_rate": 0.0002, "epoch": 5.790914066776136, "step": 5290}, {"loss": 0.3776, "grad_norm": 1.2110556364059448, "learning_rate": 0.0002, "epoch": 5.801860974274767, "step": 5300}, {"loss": 0.4049, "grad_norm": 1.3799304962158203, "learning_rate": 0.0002, "epoch": 5.812807881773399, "step": 5310}, {"loss": 0.4362, "grad_norm": 1.0570626258850098, "learning_rate": 0.0002, "epoch": 5.823754789272031, "step": 5320}, {"loss": 0.4716, "grad_norm": 1.4654436111450195, "learning_rate": 0.0002, "epoch": 5.834701696770662, "step": 5330}, {"loss": 0.4048, "grad_norm": 1.5216940641403198, "learning_rate": 0.0002, "epoch": 5.845648604269294, "step": 5340}, {"loss": 0.3848, "grad_norm": 1.018646001815796, "learning_rate": 0.0002, "epoch": 5.856595511767925, "step": 5350}, {"loss": 0.3705, "grad_norm": 1.028951644897461, "learning_rate": 0.0002, "epoch": 5.867542419266557, "step": 5360}, {"loss": 0.4213, "grad_norm": 2.571263313293457, "learning_rate": 0.0002, "epoch": 5.878489326765189, "step": 5370}, {"loss": 0.3647, "grad_norm": 1.3323984146118164, "learning_rate": 0.0002, "epoch": 5.889436234263821, "step": 5380}, {"loss": 0.4085, "grad_norm": 1.4317777156829834, "learning_rate": 0.0002, "epoch": 5.900383141762452, "step": 5390}, {"loss": 0.4254, "grad_norm": 1.4289140701293945, "learning_rate": 0.0002, "epoch": 5.911330049261084, "step": 5400}, {"loss": 0.3993, "grad_norm": 1.3130780458450317, "learning_rate": 0.0002, "epoch": 5.922276956759715, "step": 5410}, {"loss": 0.4025, "grad_norm": 1.3979902267456055, "learning_rate": 0.0002, "epoch": 5.933223864258347, "step": 5420}, {"loss": 0.3997, "grad_norm": 1.1827352046966553, "learning_rate": 0.0002, "epoch": 5.944170771756979, "step": 5430}, {"loss": 0.4163, "grad_norm": 1.1672080755233765, "learning_rate": 0.0002, "epoch": 5.95511767925561, "step": 5440}, {"loss": 0.4425, "grad_norm": 1.0949620008468628, "learning_rate": 0.0002, "epoch": 5.966064586754242, "step": 5450}, {"loss": 0.4219, "grad_norm": 1.3183925151824951, "learning_rate": 0.0002, "epoch": 5.977011494252873, "step": 5460}, {"loss": 0.4171, "grad_norm": 1.096198320388794, "learning_rate": 0.0002, "epoch": 5.987958401751505, "step": 5470}, {"loss": 0.3886, "grad_norm": 1.2601423263549805, "learning_rate": 0.0002, "epoch": 5.998905309250137, "step": 5480}, {"eval_loss": 1.611358880996704, "eval_runtime": 46.0638, "eval_samples_per_second": 9.465, "eval_steps_per_second": 1.194, "epoch": 6.0, "step": 5481}, {"loss": 0.2616, "grad_norm": 0.9854364991188049, "learning_rate": 0.0002, "epoch": 6.009852216748769, "step": 5490}, {"loss": 0.2412, "grad_norm": 1.8073689937591553, "learning_rate": 0.0002, "epoch": 6.0207991242474, "step": 5500}, {"loss": 0.2317, "grad_norm": 1.1852164268493652, "learning_rate": 0.0002, "epoch": 6.031746031746032, "step": 5510}, {"loss": 0.224, "grad_norm": 1.0937914848327637, "learning_rate": 0.0002, "epoch": 6.042692939244663, "step": 5520}, {"loss": 0.2473, "grad_norm": 0.7411194443702698, "learning_rate": 0.0002, "epoch": 6.053639846743295, "step": 5530}, {"loss": 0.2846, "grad_norm": 1.552127480506897, "learning_rate": 0.0002, "epoch": 6.064586754241927, "step": 5540}, {"loss": 0.2639, "grad_norm": 1.0465604066848755, "learning_rate": 0.0002, "epoch": 6.075533661740558, "step": 5550}, {"loss": 0.2696, "grad_norm": 1.4008121490478516, "learning_rate": 0.0002, "epoch": 6.08648056923919, "step": 5560}, {"loss": 0.3049, "grad_norm": 1.7049046754837036, "learning_rate": 0.0002, "epoch": 6.097427476737821, "step": 5570}, {"loss": 0.263, "grad_norm": 1.111151933670044, "learning_rate": 0.0002, "epoch": 6.108374384236453, "step": 5580}, {"loss": 0.2816, "grad_norm": 1.4271087646484375, "learning_rate": 0.0002, "epoch": 6.119321291735085, "step": 5590}, {"loss": 0.2878, "grad_norm": 1.3917373418807983, "learning_rate": 0.0002, "epoch": 6.130268199233717, "step": 5600}, {"loss": 0.2482, "grad_norm": 1.013689637184143, "learning_rate": 0.0002, "epoch": 6.141215106732348, "step": 5610}, {"loss": 0.2841, "grad_norm": 1.342645525932312, "learning_rate": 0.0002, "epoch": 6.15216201423098, "step": 5620}, {"loss": 0.2335, "grad_norm": 1.4480562210083008, "learning_rate": 0.0002, "epoch": 6.163108921729611, "step": 5630}, {"loss": 0.2696, "grad_norm": 1.2483175992965698, "learning_rate": 0.0002, "epoch": 6.174055829228243, "step": 5640}, {"loss": 0.2656, "grad_norm": 1.2944550514221191, "learning_rate": 0.0002, "epoch": 6.185002736726875, "step": 5650}, {"loss": 0.2704, "grad_norm": 1.264142632484436, "learning_rate": 0.0002, "epoch": 6.195949644225506, "step": 5660}, {"loss": 0.2971, "grad_norm": 1.2068781852722168, "learning_rate": 0.0002, "epoch": 6.206896551724138, "step": 5670}, {"loss": 0.2882, "grad_norm": 1.0401629209518433, "learning_rate": 0.0002, "epoch": 6.217843459222769, "step": 5680}, {"loss": 0.3022, "grad_norm": 1.2054402828216553, "learning_rate": 0.0002, "epoch": 6.228790366721401, "step": 5690}, {"loss": 0.2949, "grad_norm": 1.1278687715530396, "learning_rate": 0.0002, "epoch": 6.239737274220033, "step": 5700}, {"loss": 0.2477, "grad_norm": 1.24592125415802, "learning_rate": 0.0002, "epoch": 6.250684181718665, "step": 5710}, {"loss": 0.246, "grad_norm": 1.2686697244644165, "learning_rate": 0.0002, "epoch": 6.261631089217296, "step": 5720}, {"loss": 0.2974, "grad_norm": 1.1836518049240112, "learning_rate": 0.0002, "epoch": 6.272577996715928, "step": 5730}, {"loss": 0.2963, "grad_norm": 1.387752890586853, "learning_rate": 0.0002, "epoch": 6.283524904214559, "step": 5740}, {"loss": 0.2961, "grad_norm": 1.9390363693237305, "learning_rate": 0.0002, "epoch": 6.294471811713191, "step": 5750}, {"loss": 0.2765, "grad_norm": 1.2919824123382568, "learning_rate": 0.0002, "epoch": 6.305418719211823, "step": 5760}, {"loss": 0.2898, "grad_norm": 1.2793965339660645, "learning_rate": 0.0002, "epoch": 6.316365626710454, "step": 5770}, {"loss": 0.2786, "grad_norm": 1.5486980676651, "learning_rate": 0.0002, "epoch": 6.327312534209086, "step": 5780}, {"loss": 0.2684, "grad_norm": 1.2757408618927002, "learning_rate": 0.0002, "epoch": 6.338259441707717, "step": 5790}, {"loss": 0.2841, "grad_norm": 1.3245713710784912, "learning_rate": 0.0002, "epoch": 6.349206349206349, "step": 5800}, {"loss": 0.3096, "grad_norm": 1.6262527704238892, "learning_rate": 0.0002, "epoch": 6.360153256704981, "step": 5810}, {"loss": 0.3219, "grad_norm": 1.465224027633667, "learning_rate": 0.0002, "epoch": 6.371100164203613, "step": 5820}, {"loss": 0.2703, "grad_norm": 1.437408447265625, "learning_rate": 0.0002, "epoch": 6.382047071702244, "step": 5830}, {"loss": 0.3012, "grad_norm": 1.3094626665115356, "learning_rate": 0.0002, "epoch": 6.392993979200876, "step": 5840}, {"loss": 0.2991, "grad_norm": 1.6717544794082642, "learning_rate": 0.0002, "epoch": 6.403940886699507, "step": 5850}, {"loss": 0.2892, "grad_norm": 1.1023344993591309, "learning_rate": 0.0002, "epoch": 6.414887794198139, "step": 5860}, {"loss": 0.3078, "grad_norm": 1.2397106885910034, "learning_rate": 0.0002, "epoch": 6.425834701696771, "step": 5870}, {"loss": 0.2984, "grad_norm": 1.6139185428619385, "learning_rate": 0.0002, "epoch": 6.436781609195402, "step": 5880}, {"loss": 0.2353, "grad_norm": 1.3164576292037964, "learning_rate": 0.0002, "epoch": 6.447728516694034, "step": 5890}, {"loss": 0.2772, "grad_norm": 1.3317217826843262, "learning_rate": 0.0002, "epoch": 6.458675424192665, "step": 5900}, {"loss": 0.2555, "grad_norm": 1.215008020401001, "learning_rate": 0.0002, "epoch": 6.469622331691297, "step": 5910}, {"loss": 0.2715, "grad_norm": 1.625672698020935, "learning_rate": 0.0002, "epoch": 6.480569239189929, "step": 5920}, {"loss": 0.2938, "grad_norm": 1.1262489557266235, "learning_rate": 0.0002, "epoch": 6.491516146688561, "step": 5930}, {"loss": 0.2921, "grad_norm": 1.447100281715393, "learning_rate": 0.0002, "epoch": 6.502463054187192, "step": 5940}, {"loss": 0.3059, "grad_norm": 1.3306448459625244, "learning_rate": 0.0002, "epoch": 6.513409961685824, "step": 5950}, {"loss": 0.2922, "grad_norm": 1.307732105255127, "learning_rate": 0.0002, "epoch": 6.524356869184455, "step": 5960}, {"loss": 0.2891, "grad_norm": 1.1851097345352173, "learning_rate": 0.0002, "epoch": 6.535303776683087, "step": 5970}, {"loss": 0.2859, "grad_norm": 1.462816596031189, "learning_rate": 0.0002, "epoch": 6.546250684181719, "step": 5980}, {"loss": 0.2698, "grad_norm": 1.2324728965759277, "learning_rate": 0.0002, "epoch": 6.55719759168035, "step": 5990}, {"loss": 0.2672, "grad_norm": 1.3627429008483887, "learning_rate": 0.0002, "epoch": 6.568144499178982, "step": 6000}, {"loss": 0.3182, "grad_norm": 1.94977867603302, "learning_rate": 0.0002, "epoch": 6.579091406677613, "step": 6010}, {"loss": 0.3183, "grad_norm": 1.459844946861267, "learning_rate": 0.0002, "epoch": 6.590038314176245, "step": 6020}, {"loss": 0.3142, "grad_norm": 1.4454325437545776, "learning_rate": 0.0002, "epoch": 6.600985221674877, "step": 6030}, {"loss": 0.269, "grad_norm": 1.4245165586471558, "learning_rate": 0.0002, "epoch": 6.611932129173509, "step": 6040}, {"loss": 0.3041, "grad_norm": 1.195803165435791, "learning_rate": 0.0002, "epoch": 6.62287903667214, "step": 6050}, {"loss": 0.3075, "grad_norm": 1.3589898347854614, "learning_rate": 0.0002, "epoch": 6.633825944170772, "step": 6060}, {"loss": 0.3291, "grad_norm": 1.3488036394119263, "learning_rate": 0.0002, "epoch": 6.644772851669403, "step": 6070}, {"loss": 0.2898, "grad_norm": 1.0954102277755737, "learning_rate": 0.0002, "epoch": 6.655719759168035, "step": 6080}, {"loss": 0.3489, "grad_norm": 1.4431062936782837, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 6090}, {"loss": 0.2816, "grad_norm": 1.4387465715408325, "learning_rate": 0.0002, "epoch": 6.677613574165298, "step": 6100}, {"loss": 0.2705, "grad_norm": 1.8398990631103516, "learning_rate": 0.0002, "epoch": 6.68856048166393, "step": 6110}, {"loss": 0.3214, "grad_norm": 1.3523273468017578, "learning_rate": 0.0002, "epoch": 6.699507389162561, "step": 6120}, {"loss": 0.287, "grad_norm": 1.6326191425323486, "learning_rate": 0.0002, "epoch": 6.710454296661193, "step": 6130}, {"loss": 0.2857, "grad_norm": 1.3677960634231567, "learning_rate": 0.0002, "epoch": 6.721401204159825, "step": 6140}, {"loss": 0.3264, "grad_norm": 1.1993201971054077, "learning_rate": 0.0002, "epoch": 6.732348111658457, "step": 6150}, {"loss": 0.3071, "grad_norm": 1.1864078044891357, "learning_rate": 0.0002, "epoch": 6.743295019157088, "step": 6160}, {"loss": 0.3087, "grad_norm": 1.1625522375106812, "learning_rate": 0.0002, "epoch": 6.75424192665572, "step": 6170}, {"loss": 0.3551, "grad_norm": 1.5803234577178955, "learning_rate": 0.0002, "epoch": 6.765188834154351, "step": 6180}, {"loss": 0.3059, "grad_norm": 1.151746153831482, "learning_rate": 0.0002, "epoch": 6.776135741652983, "step": 6190}, {"loss": 0.2697, "grad_norm": 1.0727161169052124, "learning_rate": 0.0002, "epoch": 6.787082649151615, "step": 6200}, {"loss": 0.2844, "grad_norm": 1.4148162603378296, "learning_rate": 0.0002, "epoch": 6.798029556650246, "step": 6210}, {"loss": 0.3417, "grad_norm": 1.2071447372436523, "learning_rate": 0.0002, "epoch": 6.808976464148878, "step": 6220}, {"loss": 0.3066, "grad_norm": 1.3843804597854614, "learning_rate": 0.0002, "epoch": 6.819923371647509, "step": 6230}, {"loss": 0.2769, "grad_norm": 1.2490662336349487, "learning_rate": 0.0002, "epoch": 6.830870279146141, "step": 6240}, {"loss": 0.3237, "grad_norm": 1.6029689311981201, "learning_rate": 0.0002, "epoch": 6.841817186644773, "step": 6250}, {"loss": 0.3152, "grad_norm": 1.0388455390930176, "learning_rate": 0.0002, "epoch": 6.852764094143405, "step": 6260}, {"loss": 0.3026, "grad_norm": 1.3883857727050781, "learning_rate": 0.0002, "epoch": 6.863711001642036, "step": 6270}, {"loss": 0.3175, "grad_norm": 1.0500187873840332, "learning_rate": 0.0002, "epoch": 6.874657909140668, "step": 6280}, {"loss": 0.2952, "grad_norm": 1.4243487119674683, "learning_rate": 0.0002, "epoch": 6.885604816639299, "step": 6290}, {"loss": 0.2679, "grad_norm": 1.3169665336608887, "learning_rate": 0.0002, "epoch": 6.896551724137931, "step": 6300}, {"loss": 0.3291, "grad_norm": 1.5261493921279907, "learning_rate": 0.0002, "epoch": 6.907498631636563, "step": 6310}, {"loss": 0.3344, "grad_norm": 1.578403115272522, "learning_rate": 0.0002, "epoch": 6.9184455391351944, "step": 6320}, {"loss": 0.3263, "grad_norm": 1.4093263149261475, "learning_rate": 0.0002, "epoch": 6.929392446633826, "step": 6330}, {"loss": 0.3396, "grad_norm": 1.4003552198410034, "learning_rate": 0.0002, "epoch": 6.940339354132457, "step": 6340}, {"loss": 0.3476, "grad_norm": 1.650190830230713, "learning_rate": 0.0002, "epoch": 6.951286261631089, "step": 6350}, {"loss": 0.3442, "grad_norm": 1.2314515113830566, "learning_rate": 0.0002, "epoch": 6.962233169129721, "step": 6360}, {"loss": 0.3341, "grad_norm": 1.270980954170227, "learning_rate": 0.0002, "epoch": 6.973180076628353, "step": 6370}, {"loss": 0.3425, "grad_norm": 1.6352545022964478, "learning_rate": 0.0002, "epoch": 6.984126984126984, "step": 6380}, {"loss": 0.3647, "grad_norm": 1.3744925260543823, "learning_rate": 0.0002, "epoch": 6.995073891625616, "step": 6390}, {"eval_loss": 1.756764531135559, "eval_runtime": 46.0542, "eval_samples_per_second": 9.467, "eval_steps_per_second": 1.194, "epoch": 6.999452654625069, "step": 6394}, {"loss": 0.2356, "grad_norm": 0.856991171836853, "learning_rate": 0.0002, "epoch": 7.006020799124247, "step": 6400}, {"loss": 0.2138, "grad_norm": 0.9483422040939331, "learning_rate": 0.0002, "epoch": 7.016967706622879, "step": 6410}, {"loss": 0.1884, "grad_norm": 1.0703433752059937, "learning_rate": 0.0002, "epoch": 7.027914614121511, "step": 6420}, {"loss": 0.2088, "grad_norm": 1.761413812637329, "learning_rate": 0.0002, "epoch": 7.0388615216201424, "step": 6430}, {"loss": 0.2156, "grad_norm": 0.9484238028526306, "learning_rate": 0.0002, "epoch": 7.049808429118774, "step": 6440}, {"loss": 0.2323, "grad_norm": 1.5663186311721802, "learning_rate": 0.0002, "epoch": 7.060755336617405, "step": 6450}, {"loss": 0.2089, "grad_norm": 0.7692174315452576, "learning_rate": 0.0002, "epoch": 7.071702244116037, "step": 6460}, {"loss": 0.1856, "grad_norm": 1.3554800748825073, "learning_rate": 0.0002, "epoch": 7.082649151614669, "step": 6470}, {"loss": 0.2057, "grad_norm": 0.9705919027328491, "learning_rate": 0.0002, "epoch": 7.093596059113301, "step": 6480}, {"loss": 0.2068, "grad_norm": 1.355778694152832, "learning_rate": 0.0002, "epoch": 7.104542966611932, "step": 6490}, {"loss": 0.2021, "grad_norm": 1.5389477014541626, "learning_rate": 0.0002, "epoch": 7.115489874110564, "step": 6500}, {"loss": 0.1963, "grad_norm": 0.9565434455871582, "learning_rate": 0.0002, "epoch": 7.126436781609195, "step": 6510}, {"loss": 0.2323, "grad_norm": 1.101539134979248, "learning_rate": 0.0002, "epoch": 7.137383689107827, "step": 6520}, {"loss": 0.2024, "grad_norm": 0.925153374671936, "learning_rate": 0.0002, "epoch": 7.148330596606459, "step": 6530}, {"loss": 0.1955, "grad_norm": 1.1609078645706177, "learning_rate": 0.0002, "epoch": 7.1592775041050905, "step": 6540}, {"loss": 0.2022, "grad_norm": 0.8908484578132629, "learning_rate": 0.0002, "epoch": 7.170224411603722, "step": 6550}, {"loss": 0.2378, "grad_norm": 0.9066158533096313, "learning_rate": 0.0002, "epoch": 7.181171319102353, "step": 6560}, {"loss": 0.1955, "grad_norm": 1.3601553440093994, "learning_rate": 0.0002, "epoch": 7.192118226600985, "step": 6570}, {"loss": 0.241, "grad_norm": 1.0034444332122803, "learning_rate": 0.0002, "epoch": 7.203065134099617, "step": 6580}, {"loss": 0.2134, "grad_norm": 1.608299970626831, "learning_rate": 0.0002, "epoch": 7.214012041598249, "step": 6590}, {"loss": 0.2089, "grad_norm": 1.2889668941497803, "learning_rate": 0.0002, "epoch": 7.22495894909688, "step": 6600}, {"loss": 0.2405, "grad_norm": 0.9896159768104553, "learning_rate": 0.0002, "epoch": 7.235905856595512, "step": 6610}, {"loss": 0.2091, "grad_norm": 1.408511996269226, "learning_rate": 0.0002, "epoch": 7.246852764094143, "step": 6620}, {"loss": 0.1958, "grad_norm": 1.0823664665222168, "learning_rate": 0.0002, "epoch": 7.257799671592775, "step": 6630}, {"loss": 0.2117, "grad_norm": 1.027026891708374, "learning_rate": 0.0002, "epoch": 7.268746579091407, "step": 6640}, {"loss": 0.2319, "grad_norm": 1.0922648906707764, "learning_rate": 0.0002, "epoch": 7.2796934865900385, "step": 6650}, {"loss": 0.2367, "grad_norm": 1.3361082077026367, "learning_rate": 0.0002, "epoch": 7.29064039408867, "step": 6660}, {"loss": 0.2299, "grad_norm": 1.9565683603286743, "learning_rate": 0.0002, "epoch": 7.301587301587301, "step": 6670}, {"loss": 0.2248, "grad_norm": 1.413672685623169, "learning_rate": 0.0002, "epoch": 7.312534209085933, "step": 6680}, {"loss": 0.2306, "grad_norm": 1.121842384338379, "learning_rate": 0.0002, "epoch": 7.323481116584565, "step": 6690}, {"loss": 0.2222, "grad_norm": 1.0622057914733887, "learning_rate": 0.0002, "epoch": 7.334428024083197, "step": 6700}, {"loss": 0.2387, "grad_norm": 1.280921459197998, "learning_rate": 0.0002, "epoch": 7.345374931581828, "step": 6710}, {"loss": 0.2428, "grad_norm": 1.5295953750610352, "learning_rate": 0.0002, "epoch": 7.35632183908046, "step": 6720}, {"loss": 0.2149, "grad_norm": 1.4289230108261108, "learning_rate": 0.0002, "epoch": 7.367268746579091, "step": 6730}, {"loss": 0.2172, "grad_norm": 1.535111665725708, "learning_rate": 0.0002, "epoch": 7.378215654077723, "step": 6740}, {"loss": 0.2262, "grad_norm": 1.777826189994812, "learning_rate": 0.0002, "epoch": 7.389162561576355, "step": 6750}, {"loss": 0.2246, "grad_norm": 1.5058139562606812, "learning_rate": 0.0002, "epoch": 7.4001094690749865, "step": 6760}, {"loss": 0.2323, "grad_norm": 0.9381663203239441, "learning_rate": 0.0002, "epoch": 7.411056376573618, "step": 6770}, {"loss": 0.2078, "grad_norm": 1.4739434719085693, "learning_rate": 0.0002, "epoch": 7.422003284072249, "step": 6780}, {"loss": 0.2493, "grad_norm": 1.8703559637069702, "learning_rate": 0.0002, "epoch": 7.432950191570881, "step": 6790}, {"loss": 0.2554, "grad_norm": 1.2242027521133423, "learning_rate": 0.0002, "epoch": 7.443897099069513, "step": 6800}, {"loss": 0.2258, "grad_norm": 1.3950374126434326, "learning_rate": 0.0002, "epoch": 7.454844006568145, "step": 6810}, {"loss": 0.2365, "grad_norm": 1.461701512336731, "learning_rate": 0.0002, "epoch": 7.465790914066776, "step": 6820}, {"loss": 0.2302, "grad_norm": 1.4460340738296509, "learning_rate": 0.0002, "epoch": 7.476737821565408, "step": 6830}, {"loss": 0.2294, "grad_norm": 1.0341510772705078, "learning_rate": 0.0002, "epoch": 7.487684729064039, "step": 6840}, {"loss": 0.2338, "grad_norm": 0.8885145783424377, "learning_rate": 0.0002, "epoch": 7.498631636562671, "step": 6850}, {"loss": 0.2424, "grad_norm": 2.4326062202453613, "learning_rate": 0.0002, "epoch": 7.509578544061303, "step": 6860}, {"loss": 0.2352, "grad_norm": 1.1390372514724731, "learning_rate": 0.0002, "epoch": 7.5205254515599345, "step": 6870}, {"loss": 0.2184, "grad_norm": 1.2346464395523071, "learning_rate": 0.0002, "epoch": 7.531472359058566, "step": 6880}, {"loss": 0.2389, "grad_norm": 1.6705836057662964, "learning_rate": 0.0002, "epoch": 7.542419266557197, "step": 6890}, {"loss": 0.2346, "grad_norm": 0.8130379319190979, "learning_rate": 0.0002, "epoch": 7.553366174055829, "step": 6900}, {"loss": 0.2165, "grad_norm": 1.2974088191986084, "learning_rate": 0.0002, "epoch": 7.564313081554461, "step": 6910}, {"loss": 0.2328, "grad_norm": 1.3465348482131958, "learning_rate": 0.0002, "epoch": 7.575259989053093, "step": 6920}, {"loss": 0.2789, "grad_norm": 1.245126724243164, "learning_rate": 0.0002, "epoch": 7.586206896551724, "step": 6930}, {"loss": 0.2446, "grad_norm": 1.3736917972564697, "learning_rate": 0.0002, "epoch": 7.597153804050356, "step": 6940}, {"loss": 0.2483, "grad_norm": 1.340989351272583, "learning_rate": 0.0002, "epoch": 7.608100711548987, "step": 6950}, {"loss": 0.3014, "grad_norm": 1.1082850694656372, "learning_rate": 0.0002, "epoch": 7.619047619047619, "step": 6960}, {"loss": 0.2319, "grad_norm": 1.3829188346862793, "learning_rate": 0.0002, "epoch": 7.629994526546251, "step": 6970}, {"loss": 0.2299, "grad_norm": 1.5384989976882935, "learning_rate": 0.0002, "epoch": 7.6409414340448825, "step": 6980}, {"loss": 0.2457, "grad_norm": 1.1061540842056274, "learning_rate": 0.0002, "epoch": 7.651888341543514, "step": 6990}, {"loss": 0.2216, "grad_norm": 1.2673815488815308, "learning_rate": 0.0002, "epoch": 7.662835249042145, "step": 7000}, {"loss": 0.2397, "grad_norm": 1.2290737628936768, "learning_rate": 0.0002, "epoch": 7.673782156540777, "step": 7010}, {"loss": 0.2112, "grad_norm": 1.4055291414260864, "learning_rate": 0.0002, "epoch": 7.684729064039409, "step": 7020}, {"loss": 0.2548, "grad_norm": 1.7786750793457031, "learning_rate": 0.0002, "epoch": 7.695675971538041, "step": 7030}, {"loss": 0.2241, "grad_norm": 1.454209566116333, "learning_rate": 0.0002, "epoch": 7.706622879036672, "step": 7040}, {"loss": 0.2461, "grad_norm": 1.3995633125305176, "learning_rate": 0.0002, "epoch": 7.717569786535304, "step": 7050}, {"loss": 0.2785, "grad_norm": 1.7514715194702148, "learning_rate": 0.0002, "epoch": 7.728516694033935, "step": 7060}, {"loss": 0.2335, "grad_norm": 1.5538004636764526, "learning_rate": 0.0002, "epoch": 7.739463601532567, "step": 7070}, {"loss": 0.2245, "grad_norm": 1.122506856918335, "learning_rate": 0.0002, "epoch": 7.750410509031199, "step": 7080}, {"loss": 0.2473, "grad_norm": 1.2445831298828125, "learning_rate": 0.0002, "epoch": 7.7613574165298305, "step": 7090}, {"loss": 0.2421, "grad_norm": 1.1478949785232544, "learning_rate": 0.0002, "epoch": 7.772304324028462, "step": 7100}, {"loss": 0.2346, "grad_norm": 1.4352518320083618, "learning_rate": 0.0002, "epoch": 7.783251231527093, "step": 7110}, {"loss": 0.2351, "grad_norm": 1.511096715927124, "learning_rate": 0.0002, "epoch": 7.794198139025725, "step": 7120}, {"loss": 0.2586, "grad_norm": 1.2296271324157715, "learning_rate": 0.0002, "epoch": 7.805145046524357, "step": 7130}, {"loss": 0.2453, "grad_norm": 1.7886443138122559, "learning_rate": 0.0002, "epoch": 7.816091954022989, "step": 7140}, {"loss": 0.2645, "grad_norm": 1.8886322975158691, "learning_rate": 0.0002, "epoch": 7.82703886152162, "step": 7150}, {"loss": 0.2658, "grad_norm": 1.3493725061416626, "learning_rate": 0.0002, "epoch": 7.837985769020252, "step": 7160}, {"loss": 0.2787, "grad_norm": 1.379209041595459, "learning_rate": 0.0002, "epoch": 7.848932676518883, "step": 7170}, {"loss": 0.248, "grad_norm": 0.9374330043792725, "learning_rate": 0.0002, "epoch": 7.859879584017515, "step": 7180}, {"loss": 0.2412, "grad_norm": 1.0391291379928589, "learning_rate": 0.0002, "epoch": 7.870826491516147, "step": 7190}, {"loss": 0.2648, "grad_norm": 1.2710281610488892, "learning_rate": 0.0002, "epoch": 7.8817733990147785, "step": 7200}, {"loss": 0.2511, "grad_norm": 1.6858662366867065, "learning_rate": 0.0002, "epoch": 7.89272030651341, "step": 7210}, {"loss": 0.2667, "grad_norm": 1.0925853252410889, "learning_rate": 0.0002, "epoch": 7.903667214012041, "step": 7220}, {"loss": 0.2756, "grad_norm": 1.7404073476791382, "learning_rate": 0.0002, "epoch": 7.914614121510673, "step": 7230}, {"loss": 0.2486, "grad_norm": 1.284067153930664, "learning_rate": 0.0002, "epoch": 7.925561029009305, "step": 7240}, {"loss": 0.2394, "grad_norm": 1.3801543712615967, "learning_rate": 0.0002, "epoch": 7.936507936507937, "step": 7250}, {"loss": 0.2761, "grad_norm": 1.4068974256515503, "learning_rate": 0.0002, "epoch": 7.947454844006568, "step": 7260}, {"loss": 0.2455, "grad_norm": 1.770037055015564, "learning_rate": 0.0002, "epoch": 7.9584017515052, "step": 7270}, {"loss": 0.2863, "grad_norm": 1.473775029182434, "learning_rate": 0.0002, "epoch": 7.969348659003831, "step": 7280}, {"loss": 0.2722, "grad_norm": 1.4878343343734741, "learning_rate": 0.0002, "epoch": 7.980295566502463, "step": 7290}, {"loss": 0.2921, "grad_norm": 1.2178987264633179, "learning_rate": 0.0002, "epoch": 7.991242474001095, "step": 7300}]}