diff --git a/.gitattributes b/.gitattributes
index 5ecdab0915a1d5a508ce4edd5c306de850e1ac1f..e7fddcd5fc197e678828bc6b231f93bf39d3c900 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -1534,3 +1534,12 @@ gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-617-sd-4/checkpoint-496/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-617-sd-4/checkpoint-62/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.6-num-617-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..880901fe41029fff54981e67b19f9562a7d4ebdd
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f63771989885dd3030480c888f982b482a4f0b6609b4f6205f6bf9b1faaccaf
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..880901fe41029fff54981e67b19f9562a7d4ebdd
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f63771989885dd3030480c888f982b482a4f0b6609b4f6205f6bf9b1faaccaf
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..deef6ae3165558586e01373aa39ec4c0c4b773b5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a14bd7bd809dcf34e3681aa306e3332e2ce2c7c355217ad047c7a5749bf95e6
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0e1b09a5dbcfe048eb44a413857b39f43d94d5e0
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c4eb9f332793c665c686d8296af92d4a1ded65117b574b2bd8d775e5ba49b28
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..08608f46a39eb77cc111483d0da7109bea8c5c76
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:660ab54d0c298f22afda4a2655936aecd813daf93f4fc040b7d285b3684be4c0
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a87f6d30d6fd3f89708ffc32c0c21f542f80d156
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/trainer_state.json
@@ -0,0 +1,755 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 0.9995129079396006,
+  "eval_steps": 10,
+  "global_step": 1026,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 5.272488488730624e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..12ef772eb1159ae731240b0f1e1856ca61f42cea
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3a48a7bbe9dd3ba0a69f9915246bc8c09d59eb48fa97982bb17ecae6a0e4574
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ca3b9296d0df17d668eb413821dfcd30d49fbb9e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3561b40d2f24fb85b054ab9b43e627a22934108af1a380436def56674ca11441
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4921bc38e6acd53255e0202b75480b41356aaf87
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ceaa5adcb527c2f297d40f7bfac956b3d7389765453573b67579e3b48bff060c
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1c3b9bed50751c13ee5b8e48c5fc22e572d60c34
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26ebded2a9aa7f2379ce88e70901f21f9672c01c5b0e90fa087e6a51f5eb1d78
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..d4dcc27d0929dcf8f5c9668d87e2099c83dc75c6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/trainer_state.json
@@ -0,0 +1,1484 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2053,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 1.0544976977461248e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-2053/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..65512bc6f3fc67be900be6e35a62a1e7559f14f7
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb5e11916d1dac910d80aeb5e2e29db4a82ddf794e374a22fe3426b5f5037067
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..04141e14e46c60fdcd9cc06b6c7f1bd83814ffd6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c930d03d2d48904855c717f28e9f06f0cb2b65b44e88d53f235435c21b001df
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..487e81911a510e49fb7a86bcbacbf01e4556acc7
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d766ed32a27c4226af4dbeab8fff7fa52d389f2c0263d7c8cc30e1892c1893e
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ae86a8a175903eaea7051749a24e8d86d3be8b2d
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:024b0220d9692f849dabe772d5f3ffd009410a23eae736dd4e3dfd454a3e41f0
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7398c82d8e9f11607983783ceaa89c997c9a1ab
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/trainer_state.json
@@ -0,0 +1,2206 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 2.9995129079396006,
+  "eval_steps": 10,
+  "global_step": 3079,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 1.5817465466191872e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-3079/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9a1b673a485c6d49b2f1c19fca7dcff22c7519fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eecae5ade18bceb41c2e2d950888a3c5fc7c7eb9b22f7a79be673d57aeebc0cb
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..feb097f02a656db7a8afdfa9cf15ef42b4f89fe6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a02776176843d555f9ac137f9ee80709af7c0bc6d6bd635d984a5ad01b6d8f19
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..363a8cd676b3c311e6111610f15fb2adcb6763c5
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2964ed1ed02c299e8f77f89417ef45f32346df96327ad950efe91cc89ca0ef0
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..39e10b564e4ee08c5e32273c2e11f793fb847cf6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be842ef516859975dd85baf06d955a7d7408a2bc63646e1b3e2cff0eba1c3164
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9dba95477b6a4a28f5f4583e87319d4dbae3ca0b
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/trainer_state.json
@@ -0,0 +1,2935 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 4106,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    },
+    {
+      "epoch": 3.0004870920603994,
+      "grad_norm": 0.49992576241493225,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 3.0102289332683876,
+      "grad_norm": 0.8242783546447754,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3090
+    },
+    {
+      "epoch": 3.019970774476376,
+      "grad_norm": 0.6330569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3100
+    },
+    {
+      "epoch": 3.0297126156843643,
+      "grad_norm": 0.566097617149353,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 3110
+    },
+    {
+      "epoch": 3.0394544568923525,
+      "grad_norm": 0.6337586045265198,
+      "learning_rate": 0.0002,
+      "loss": 1.4365,
+      "step": 3120
+    },
+    {
+      "epoch": 3.049196298100341,
+      "grad_norm": 0.7339403033256531,
+      "learning_rate": 0.0002,
+      "loss": 1.3916,
+      "step": 3130
+    },
+    {
+      "epoch": 3.0589381393083293,
+      "grad_norm": 0.7187346816062927,
+      "learning_rate": 0.0002,
+      "loss": 1.4617,
+      "step": 3140
+    },
+    {
+      "epoch": 3.0686799805163174,
+      "grad_norm": 0.7116255760192871,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 3150
+    },
+    {
+      "epoch": 3.078421821724306,
+      "grad_norm": 0.6493807435035706,
+      "learning_rate": 0.0002,
+      "loss": 1.4452,
+      "step": 3160
+    },
+    {
+      "epoch": 3.088163662932294,
+      "grad_norm": 0.6777266263961792,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 3170
+    },
+    {
+      "epoch": 3.0979055041402823,
+      "grad_norm": 0.6342006325721741,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 3180
+    },
+    {
+      "epoch": 3.107647345348271,
+      "grad_norm": 0.6608964204788208,
+      "learning_rate": 0.0002,
+      "loss": 1.4748,
+      "step": 3190
+    },
+    {
+      "epoch": 3.117389186556259,
+      "grad_norm": 0.7230247259140015,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 3200
+    },
+    {
+      "epoch": 3.1271310277642472,
+      "grad_norm": 0.650368332862854,
+      "learning_rate": 0.0002,
+      "loss": 1.4049,
+      "step": 3210
+    },
+    {
+      "epoch": 3.136872868972236,
+      "grad_norm": 0.7319342494010925,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 3220
+    },
+    {
+      "epoch": 3.146614710180224,
+      "grad_norm": 0.7159963846206665,
+      "learning_rate": 0.0002,
+      "loss": 1.3872,
+      "step": 3230
+    },
+    {
+      "epoch": 3.156356551388212,
+      "grad_norm": 0.8905230164527893,
+      "learning_rate": 0.0002,
+      "loss": 1.5076,
+      "step": 3240
+    },
+    {
+      "epoch": 3.1660983925962007,
+      "grad_norm": 0.6920804381370544,
+      "learning_rate": 0.0002,
+      "loss": 1.3161,
+      "step": 3250
+    },
+    {
+      "epoch": 3.175840233804189,
+      "grad_norm": 0.6782063841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.3786,
+      "step": 3260
+    },
+    {
+      "epoch": 3.1855820750121775,
+      "grad_norm": 0.735325276851654,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 3270
+    },
+    {
+      "epoch": 3.1953239162201656,
+      "grad_norm": 0.6657978296279907,
+      "learning_rate": 0.0002,
+      "loss": 1.4027,
+      "step": 3280
+    },
+    {
+      "epoch": 3.205065757428154,
+      "grad_norm": 0.771315336227417,
+      "learning_rate": 0.0002,
+      "loss": 1.3456,
+      "step": 3290
+    },
+    {
+      "epoch": 3.2148075986361424,
+      "grad_norm": 0.6492983102798462,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 3300
+    },
+    {
+      "epoch": 3.2245494398441306,
+      "grad_norm": 0.7513770461082458,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3310
+    },
+    {
+      "epoch": 3.2342912810521187,
+      "grad_norm": 0.7091423869132996,
+      "learning_rate": 0.0002,
+      "loss": 1.4032,
+      "step": 3320
+    },
+    {
+      "epoch": 3.2440331222601073,
+      "grad_norm": 0.6663975119590759,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 3330
+    },
+    {
+      "epoch": 3.2537749634680955,
+      "grad_norm": 0.6813122034072876,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 3340
+    },
+    {
+      "epoch": 3.2635168046760836,
+      "grad_norm": 0.6602569818496704,
+      "learning_rate": 0.0002,
+      "loss": 1.3681,
+      "step": 3350
+    },
+    {
+      "epoch": 3.2732586458840722,
+      "grad_norm": 0.718270480632782,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 3360
+    },
+    {
+      "epoch": 3.2830004870920604,
+      "grad_norm": 0.6884173154830933,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 3370
+    },
+    {
+      "epoch": 3.2927423283000485,
+      "grad_norm": 0.7039775848388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3380
+    },
+    {
+      "epoch": 3.302484169508037,
+      "grad_norm": 0.7444299459457397,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3390
+    },
+    {
+      "epoch": 3.3122260107160253,
+      "grad_norm": 0.7187064290046692,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 3400
+    },
+    {
+      "epoch": 3.3219678519240134,
+      "grad_norm": 0.599396288394928,
+      "learning_rate": 0.0002,
+      "loss": 1.3684,
+      "step": 3410
+    },
+    {
+      "epoch": 3.331709693132002,
+      "grad_norm": 0.7670390009880066,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3420
+    },
+    {
+      "epoch": 3.34145153433999,
+      "grad_norm": 0.6654478311538696,
+      "learning_rate": 0.0002,
+      "loss": 1.4411,
+      "step": 3430
+    },
+    {
+      "epoch": 3.351193375547979,
+      "grad_norm": 0.6644385457038879,
+      "learning_rate": 0.0002,
+      "loss": 1.4257,
+      "step": 3440
+    },
+    {
+      "epoch": 3.360935216755967,
+      "grad_norm": 0.6974098086357117,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3450
+    },
+    {
+      "epoch": 3.370677057963955,
+      "grad_norm": 0.7350399494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 3460
+    },
+    {
+      "epoch": 3.3804188991719437,
+      "grad_norm": 0.714721143245697,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 3470
+    },
+    {
+      "epoch": 3.390160740379932,
+      "grad_norm": 0.7006027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 3480
+    },
+    {
+      "epoch": 3.39990258158792,
+      "grad_norm": 0.6767925024032593,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 3490
+    },
+    {
+      "epoch": 3.4096444227959086,
+      "grad_norm": 0.6721355319023132,
+      "learning_rate": 0.0002,
+      "loss": 1.4116,
+      "step": 3500
+    },
+    {
+      "epoch": 3.419386264003897,
+      "grad_norm": 0.6845725178718567,
+      "learning_rate": 0.0002,
+      "loss": 1.443,
+      "step": 3510
+    },
+    {
+      "epoch": 3.429128105211885,
+      "grad_norm": 0.6882196664810181,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 3520
+    },
+    {
+      "epoch": 3.4388699464198735,
+      "grad_norm": 0.7663240432739258,
+      "learning_rate": 0.0002,
+      "loss": 1.4962,
+      "step": 3530
+    },
+    {
+      "epoch": 3.4486117876278617,
+      "grad_norm": 0.6304219365119934,
+      "learning_rate": 0.0002,
+      "loss": 1.4644,
+      "step": 3540
+    },
+    {
+      "epoch": 3.45835362883585,
+      "grad_norm": 0.668678879737854,
+      "learning_rate": 0.0002,
+      "loss": 1.4918,
+      "step": 3550
+    },
+    {
+      "epoch": 3.4680954700438384,
+      "grad_norm": 0.7526912093162537,
+      "learning_rate": 0.0002,
+      "loss": 1.4874,
+      "step": 3560
+    },
+    {
+      "epoch": 3.4778373112518266,
+      "grad_norm": 1.089495301246643,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 3570
+    },
+    {
+      "epoch": 3.4875791524598148,
+      "grad_norm": 0.7282902002334595,
+      "learning_rate": 0.0002,
+      "loss": 1.3871,
+      "step": 3580
+    },
+    {
+      "epoch": 3.4973209936678034,
+      "grad_norm": 0.6540156602859497,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3590
+    },
+    {
+      "epoch": 3.5070628348757915,
+      "grad_norm": 0.6449568867683411,
+      "learning_rate": 0.0002,
+      "loss": 1.4367,
+      "step": 3600
+    },
+    {
+      "epoch": 3.5168046760837797,
+      "grad_norm": 0.7262216210365295,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 3610
+    },
+    {
+      "epoch": 3.5265465172917683,
+      "grad_norm": 0.6048615574836731,
+      "learning_rate": 0.0002,
+      "loss": 1.4374,
+      "step": 3620
+    },
+    {
+      "epoch": 3.5362883584997564,
+      "grad_norm": 0.6780537366867065,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 3630
+    },
+    {
+      "epoch": 3.5460301997077446,
+      "grad_norm": 0.6851925253868103,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 3640
+    },
+    {
+      "epoch": 3.555772040915733,
+      "grad_norm": 0.6530634164810181,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 3650
+    },
+    {
+      "epoch": 3.5655138821237213,
+      "grad_norm": 0.7193992733955383,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 3660
+    },
+    {
+      "epoch": 3.5752557233317095,
+      "grad_norm": 0.767496645450592,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 3670
+    },
+    {
+      "epoch": 3.584997564539698,
+      "grad_norm": 0.6912919282913208,
+      "learning_rate": 0.0002,
+      "loss": 1.4824,
+      "step": 3680
+    },
+    {
+      "epoch": 3.5947394057476862,
+      "grad_norm": 0.7383436560630798,
+      "learning_rate": 0.0002,
+      "loss": 1.4497,
+      "step": 3690
+    },
+    {
+      "epoch": 3.6044812469556744,
+      "grad_norm": 0.6746662855148315,
+      "learning_rate": 0.0002,
+      "loss": 1.4822,
+      "step": 3700
+    },
+    {
+      "epoch": 3.614223088163663,
+      "grad_norm": 0.6885138750076294,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 3710
+    },
+    {
+      "epoch": 3.623964929371651,
+      "grad_norm": 0.6694392561912537,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 3720
+    },
+    {
+      "epoch": 3.6337067705796393,
+      "grad_norm": 0.812358021736145,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 3730
+    },
+    {
+      "epoch": 3.643448611787628,
+      "grad_norm": 0.7267130017280579,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 3740
+    },
+    {
+      "epoch": 3.653190452995616,
+      "grad_norm": 0.6958749294281006,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 3750
+    },
+    {
+      "epoch": 3.6629322942036042,
+      "grad_norm": 0.6805673241615295,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3760
+    },
+    {
+      "epoch": 3.672674135411593,
+      "grad_norm": 0.7184410095214844,
+      "learning_rate": 0.0002,
+      "loss": 1.4338,
+      "step": 3770
+    },
+    {
+      "epoch": 3.682415976619581,
+      "grad_norm": 0.7716330289840698,
+      "learning_rate": 0.0002,
+      "loss": 1.3935,
+      "step": 3780
+    },
+    {
+      "epoch": 3.6921578178275696,
+      "grad_norm": 0.6675831079483032,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 3790
+    },
+    {
+      "epoch": 3.7018996590355577,
+      "grad_norm": 0.6480095386505127,
+      "learning_rate": 0.0002,
+      "loss": 1.401,
+      "step": 3800
+    },
+    {
+      "epoch": 3.711641500243546,
+      "grad_norm": 0.6559418439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 3810
+    },
+    {
+      "epoch": 3.7213833414515345,
+      "grad_norm": 0.6596545577049255,
+      "learning_rate": 0.0002,
+      "loss": 1.4341,
+      "step": 3820
+    },
+    {
+      "epoch": 3.7311251826595226,
+      "grad_norm": 0.7172950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3830
+    },
+    {
+      "epoch": 3.740867023867511,
+      "grad_norm": 0.796148419380188,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 3840
+    },
+    {
+      "epoch": 3.7506088650754994,
+      "grad_norm": 0.6600322723388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4992,
+      "step": 3850
+    },
+    {
+      "epoch": 3.7603507062834876,
+      "grad_norm": 0.6776387691497803,
+      "learning_rate": 0.0002,
+      "loss": 1.4201,
+      "step": 3860
+    },
+    {
+      "epoch": 3.770092547491476,
+      "grad_norm": 0.7768304347991943,
+      "learning_rate": 0.0002,
+      "loss": 1.3893,
+      "step": 3870
+    },
+    {
+      "epoch": 3.7798343886994643,
+      "grad_norm": 1.0579794645309448,
+      "learning_rate": 0.0002,
+      "loss": 1.4886,
+      "step": 3880
+    },
+    {
+      "epoch": 3.7895762299074525,
+      "grad_norm": 0.6757252812385559,
+      "learning_rate": 0.0002,
+      "loss": 1.4556,
+      "step": 3890
+    },
+    {
+      "epoch": 3.799318071115441,
+      "grad_norm": 0.6706996560096741,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 3900
+    },
+    {
+      "epoch": 3.809059912323429,
+      "grad_norm": 0.7026948928833008,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 3910
+    },
+    {
+      "epoch": 3.8188017535314174,
+      "grad_norm": 0.6437768340110779,
+      "learning_rate": 0.0002,
+      "loss": 1.5487,
+      "step": 3920
+    },
+    {
+      "epoch": 3.828543594739406,
+      "grad_norm": 0.7015706300735474,
+      "learning_rate": 0.0002,
+      "loss": 1.4678,
+      "step": 3930
+    },
+    {
+      "epoch": 3.838285435947394,
+      "grad_norm": 0.7049482464790344,
+      "learning_rate": 0.0002,
+      "loss": 1.4891,
+      "step": 3940
+    },
+    {
+      "epoch": 3.8480272771553823,
+      "grad_norm": 0.6533724665641785,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 3950
+    },
+    {
+      "epoch": 3.857769118363371,
+      "grad_norm": 0.7312499284744263,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 3960
+    },
+    {
+      "epoch": 3.867510959571359,
+      "grad_norm": 0.6858801245689392,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3970
+    },
+    {
+      "epoch": 3.877252800779347,
+      "grad_norm": 0.770423173904419,
+      "learning_rate": 0.0002,
+      "loss": 1.4423,
+      "step": 3980
+    },
+    {
+      "epoch": 3.886994641987336,
+      "grad_norm": 0.6987539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 3990
+    },
+    {
+      "epoch": 3.896736483195324,
+      "grad_norm": 0.7072722315788269,
+      "learning_rate": 0.0002,
+      "loss": 1.4791,
+      "step": 4000
+    },
+    {
+      "epoch": 3.906478324403312,
+      "grad_norm": 0.6492931842803955,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4010
+    },
+    {
+      "epoch": 3.9162201656113007,
+      "grad_norm": 0.7716232538223267,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 4020
+    },
+    {
+      "epoch": 3.925962006819289,
+      "grad_norm": 0.722949743270874,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 4030
+    },
+    {
+      "epoch": 3.935703848027277,
+      "grad_norm": 0.7434365749359131,
+      "learning_rate": 0.0002,
+      "loss": 1.3914,
+      "step": 4040
+    },
+    {
+      "epoch": 3.9454456892352656,
+      "grad_norm": 0.6691509485244751,
+      "learning_rate": 0.0002,
+      "loss": 1.4763,
+      "step": 4050
+    },
+    {
+      "epoch": 3.9551875304432538,
+      "grad_norm": 0.6850284337997437,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 4060
+    },
+    {
+      "epoch": 3.964929371651242,
+      "grad_norm": 0.6954452991485596,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4070
+    },
+    {
+      "epoch": 3.9746712128592305,
+      "grad_norm": 0.9316364526748657,
+      "learning_rate": 0.0002,
+      "loss": 1.417,
+      "step": 4080
+    },
+    {
+      "epoch": 3.9844130540672187,
+      "grad_norm": 0.6908289194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 4090
+    },
+    {
+      "epoch": 3.994154895275207,
+      "grad_norm": 0.666782021522522,
+      "learning_rate": 0.0002,
+      "loss": 1.4404,
+      "step": 4100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9233275651931763,
+      "eval_runtime": 55.9536,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 1.144,
+      "step": 4106
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 2.1089953954922496e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-4106/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b487f7049f5f340d2abdc9a44162faf6a13b70b9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9468b2f6834f3c73875087280b883ab24849e7b2b49ef8c169afbe03f80f5bcc
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5690f8f751d22e5100cb193b6be5590f574b458c
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e82d4f36b5067df6d0c71805aba92215fb1c159bbcde1d015e7bc6da17a4128e
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..269e4a936375ac7abfdf34c6924ac370d03a5531
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3ce99a0a69cc8a4b3b9d68161778aab5768b17b0e4746fc85ca7b8d630056f4
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0102d21fdb3705adf3dfdbf03459433cceb205d6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:450ddc08f4bb85710f00f5916b9609332f1a6e377e9adbadf971157424df885f
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3cb8b1599d11a61b9e38eb84d83cf9f571ad064f
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/trainer_state.json
@@ -0,0 +1,3664 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 4.9995129079396,
+  "eval_steps": 10,
+  "global_step": 5132,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    },
+    {
+      "epoch": 3.0004870920603994,
+      "grad_norm": 0.49992576241493225,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 3.0102289332683876,
+      "grad_norm": 0.8242783546447754,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3090
+    },
+    {
+      "epoch": 3.019970774476376,
+      "grad_norm": 0.6330569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3100
+    },
+    {
+      "epoch": 3.0297126156843643,
+      "grad_norm": 0.566097617149353,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 3110
+    },
+    {
+      "epoch": 3.0394544568923525,
+      "grad_norm": 0.6337586045265198,
+      "learning_rate": 0.0002,
+      "loss": 1.4365,
+      "step": 3120
+    },
+    {
+      "epoch": 3.049196298100341,
+      "grad_norm": 0.7339403033256531,
+      "learning_rate": 0.0002,
+      "loss": 1.3916,
+      "step": 3130
+    },
+    {
+      "epoch": 3.0589381393083293,
+      "grad_norm": 0.7187346816062927,
+      "learning_rate": 0.0002,
+      "loss": 1.4617,
+      "step": 3140
+    },
+    {
+      "epoch": 3.0686799805163174,
+      "grad_norm": 0.7116255760192871,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 3150
+    },
+    {
+      "epoch": 3.078421821724306,
+      "grad_norm": 0.6493807435035706,
+      "learning_rate": 0.0002,
+      "loss": 1.4452,
+      "step": 3160
+    },
+    {
+      "epoch": 3.088163662932294,
+      "grad_norm": 0.6777266263961792,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 3170
+    },
+    {
+      "epoch": 3.0979055041402823,
+      "grad_norm": 0.6342006325721741,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 3180
+    },
+    {
+      "epoch": 3.107647345348271,
+      "grad_norm": 0.6608964204788208,
+      "learning_rate": 0.0002,
+      "loss": 1.4748,
+      "step": 3190
+    },
+    {
+      "epoch": 3.117389186556259,
+      "grad_norm": 0.7230247259140015,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 3200
+    },
+    {
+      "epoch": 3.1271310277642472,
+      "grad_norm": 0.650368332862854,
+      "learning_rate": 0.0002,
+      "loss": 1.4049,
+      "step": 3210
+    },
+    {
+      "epoch": 3.136872868972236,
+      "grad_norm": 0.7319342494010925,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 3220
+    },
+    {
+      "epoch": 3.146614710180224,
+      "grad_norm": 0.7159963846206665,
+      "learning_rate": 0.0002,
+      "loss": 1.3872,
+      "step": 3230
+    },
+    {
+      "epoch": 3.156356551388212,
+      "grad_norm": 0.8905230164527893,
+      "learning_rate": 0.0002,
+      "loss": 1.5076,
+      "step": 3240
+    },
+    {
+      "epoch": 3.1660983925962007,
+      "grad_norm": 0.6920804381370544,
+      "learning_rate": 0.0002,
+      "loss": 1.3161,
+      "step": 3250
+    },
+    {
+      "epoch": 3.175840233804189,
+      "grad_norm": 0.6782063841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.3786,
+      "step": 3260
+    },
+    {
+      "epoch": 3.1855820750121775,
+      "grad_norm": 0.735325276851654,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 3270
+    },
+    {
+      "epoch": 3.1953239162201656,
+      "grad_norm": 0.6657978296279907,
+      "learning_rate": 0.0002,
+      "loss": 1.4027,
+      "step": 3280
+    },
+    {
+      "epoch": 3.205065757428154,
+      "grad_norm": 0.771315336227417,
+      "learning_rate": 0.0002,
+      "loss": 1.3456,
+      "step": 3290
+    },
+    {
+      "epoch": 3.2148075986361424,
+      "grad_norm": 0.6492983102798462,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 3300
+    },
+    {
+      "epoch": 3.2245494398441306,
+      "grad_norm": 0.7513770461082458,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3310
+    },
+    {
+      "epoch": 3.2342912810521187,
+      "grad_norm": 0.7091423869132996,
+      "learning_rate": 0.0002,
+      "loss": 1.4032,
+      "step": 3320
+    },
+    {
+      "epoch": 3.2440331222601073,
+      "grad_norm": 0.6663975119590759,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 3330
+    },
+    {
+      "epoch": 3.2537749634680955,
+      "grad_norm": 0.6813122034072876,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 3340
+    },
+    {
+      "epoch": 3.2635168046760836,
+      "grad_norm": 0.6602569818496704,
+      "learning_rate": 0.0002,
+      "loss": 1.3681,
+      "step": 3350
+    },
+    {
+      "epoch": 3.2732586458840722,
+      "grad_norm": 0.718270480632782,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 3360
+    },
+    {
+      "epoch": 3.2830004870920604,
+      "grad_norm": 0.6884173154830933,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 3370
+    },
+    {
+      "epoch": 3.2927423283000485,
+      "grad_norm": 0.7039775848388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3380
+    },
+    {
+      "epoch": 3.302484169508037,
+      "grad_norm": 0.7444299459457397,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3390
+    },
+    {
+      "epoch": 3.3122260107160253,
+      "grad_norm": 0.7187064290046692,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 3400
+    },
+    {
+      "epoch": 3.3219678519240134,
+      "grad_norm": 0.599396288394928,
+      "learning_rate": 0.0002,
+      "loss": 1.3684,
+      "step": 3410
+    },
+    {
+      "epoch": 3.331709693132002,
+      "grad_norm": 0.7670390009880066,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3420
+    },
+    {
+      "epoch": 3.34145153433999,
+      "grad_norm": 0.6654478311538696,
+      "learning_rate": 0.0002,
+      "loss": 1.4411,
+      "step": 3430
+    },
+    {
+      "epoch": 3.351193375547979,
+      "grad_norm": 0.6644385457038879,
+      "learning_rate": 0.0002,
+      "loss": 1.4257,
+      "step": 3440
+    },
+    {
+      "epoch": 3.360935216755967,
+      "grad_norm": 0.6974098086357117,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3450
+    },
+    {
+      "epoch": 3.370677057963955,
+      "grad_norm": 0.7350399494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 3460
+    },
+    {
+      "epoch": 3.3804188991719437,
+      "grad_norm": 0.714721143245697,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 3470
+    },
+    {
+      "epoch": 3.390160740379932,
+      "grad_norm": 0.7006027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 3480
+    },
+    {
+      "epoch": 3.39990258158792,
+      "grad_norm": 0.6767925024032593,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 3490
+    },
+    {
+      "epoch": 3.4096444227959086,
+      "grad_norm": 0.6721355319023132,
+      "learning_rate": 0.0002,
+      "loss": 1.4116,
+      "step": 3500
+    },
+    {
+      "epoch": 3.419386264003897,
+      "grad_norm": 0.6845725178718567,
+      "learning_rate": 0.0002,
+      "loss": 1.443,
+      "step": 3510
+    },
+    {
+      "epoch": 3.429128105211885,
+      "grad_norm": 0.6882196664810181,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 3520
+    },
+    {
+      "epoch": 3.4388699464198735,
+      "grad_norm": 0.7663240432739258,
+      "learning_rate": 0.0002,
+      "loss": 1.4962,
+      "step": 3530
+    },
+    {
+      "epoch": 3.4486117876278617,
+      "grad_norm": 0.6304219365119934,
+      "learning_rate": 0.0002,
+      "loss": 1.4644,
+      "step": 3540
+    },
+    {
+      "epoch": 3.45835362883585,
+      "grad_norm": 0.668678879737854,
+      "learning_rate": 0.0002,
+      "loss": 1.4918,
+      "step": 3550
+    },
+    {
+      "epoch": 3.4680954700438384,
+      "grad_norm": 0.7526912093162537,
+      "learning_rate": 0.0002,
+      "loss": 1.4874,
+      "step": 3560
+    },
+    {
+      "epoch": 3.4778373112518266,
+      "grad_norm": 1.089495301246643,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 3570
+    },
+    {
+      "epoch": 3.4875791524598148,
+      "grad_norm": 0.7282902002334595,
+      "learning_rate": 0.0002,
+      "loss": 1.3871,
+      "step": 3580
+    },
+    {
+      "epoch": 3.4973209936678034,
+      "grad_norm": 0.6540156602859497,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3590
+    },
+    {
+      "epoch": 3.5070628348757915,
+      "grad_norm": 0.6449568867683411,
+      "learning_rate": 0.0002,
+      "loss": 1.4367,
+      "step": 3600
+    },
+    {
+      "epoch": 3.5168046760837797,
+      "grad_norm": 0.7262216210365295,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 3610
+    },
+    {
+      "epoch": 3.5265465172917683,
+      "grad_norm": 0.6048615574836731,
+      "learning_rate": 0.0002,
+      "loss": 1.4374,
+      "step": 3620
+    },
+    {
+      "epoch": 3.5362883584997564,
+      "grad_norm": 0.6780537366867065,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 3630
+    },
+    {
+      "epoch": 3.5460301997077446,
+      "grad_norm": 0.6851925253868103,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 3640
+    },
+    {
+      "epoch": 3.555772040915733,
+      "grad_norm": 0.6530634164810181,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 3650
+    },
+    {
+      "epoch": 3.5655138821237213,
+      "grad_norm": 0.7193992733955383,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 3660
+    },
+    {
+      "epoch": 3.5752557233317095,
+      "grad_norm": 0.767496645450592,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 3670
+    },
+    {
+      "epoch": 3.584997564539698,
+      "grad_norm": 0.6912919282913208,
+      "learning_rate": 0.0002,
+      "loss": 1.4824,
+      "step": 3680
+    },
+    {
+      "epoch": 3.5947394057476862,
+      "grad_norm": 0.7383436560630798,
+      "learning_rate": 0.0002,
+      "loss": 1.4497,
+      "step": 3690
+    },
+    {
+      "epoch": 3.6044812469556744,
+      "grad_norm": 0.6746662855148315,
+      "learning_rate": 0.0002,
+      "loss": 1.4822,
+      "step": 3700
+    },
+    {
+      "epoch": 3.614223088163663,
+      "grad_norm": 0.6885138750076294,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 3710
+    },
+    {
+      "epoch": 3.623964929371651,
+      "grad_norm": 0.6694392561912537,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 3720
+    },
+    {
+      "epoch": 3.6337067705796393,
+      "grad_norm": 0.812358021736145,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 3730
+    },
+    {
+      "epoch": 3.643448611787628,
+      "grad_norm": 0.7267130017280579,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 3740
+    },
+    {
+      "epoch": 3.653190452995616,
+      "grad_norm": 0.6958749294281006,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 3750
+    },
+    {
+      "epoch": 3.6629322942036042,
+      "grad_norm": 0.6805673241615295,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3760
+    },
+    {
+      "epoch": 3.672674135411593,
+      "grad_norm": 0.7184410095214844,
+      "learning_rate": 0.0002,
+      "loss": 1.4338,
+      "step": 3770
+    },
+    {
+      "epoch": 3.682415976619581,
+      "grad_norm": 0.7716330289840698,
+      "learning_rate": 0.0002,
+      "loss": 1.3935,
+      "step": 3780
+    },
+    {
+      "epoch": 3.6921578178275696,
+      "grad_norm": 0.6675831079483032,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 3790
+    },
+    {
+      "epoch": 3.7018996590355577,
+      "grad_norm": 0.6480095386505127,
+      "learning_rate": 0.0002,
+      "loss": 1.401,
+      "step": 3800
+    },
+    {
+      "epoch": 3.711641500243546,
+      "grad_norm": 0.6559418439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 3810
+    },
+    {
+      "epoch": 3.7213833414515345,
+      "grad_norm": 0.6596545577049255,
+      "learning_rate": 0.0002,
+      "loss": 1.4341,
+      "step": 3820
+    },
+    {
+      "epoch": 3.7311251826595226,
+      "grad_norm": 0.7172950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3830
+    },
+    {
+      "epoch": 3.740867023867511,
+      "grad_norm": 0.796148419380188,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 3840
+    },
+    {
+      "epoch": 3.7506088650754994,
+      "grad_norm": 0.6600322723388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4992,
+      "step": 3850
+    },
+    {
+      "epoch": 3.7603507062834876,
+      "grad_norm": 0.6776387691497803,
+      "learning_rate": 0.0002,
+      "loss": 1.4201,
+      "step": 3860
+    },
+    {
+      "epoch": 3.770092547491476,
+      "grad_norm": 0.7768304347991943,
+      "learning_rate": 0.0002,
+      "loss": 1.3893,
+      "step": 3870
+    },
+    {
+      "epoch": 3.7798343886994643,
+      "grad_norm": 1.0579794645309448,
+      "learning_rate": 0.0002,
+      "loss": 1.4886,
+      "step": 3880
+    },
+    {
+      "epoch": 3.7895762299074525,
+      "grad_norm": 0.6757252812385559,
+      "learning_rate": 0.0002,
+      "loss": 1.4556,
+      "step": 3890
+    },
+    {
+      "epoch": 3.799318071115441,
+      "grad_norm": 0.6706996560096741,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 3900
+    },
+    {
+      "epoch": 3.809059912323429,
+      "grad_norm": 0.7026948928833008,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 3910
+    },
+    {
+      "epoch": 3.8188017535314174,
+      "grad_norm": 0.6437768340110779,
+      "learning_rate": 0.0002,
+      "loss": 1.5487,
+      "step": 3920
+    },
+    {
+      "epoch": 3.828543594739406,
+      "grad_norm": 0.7015706300735474,
+      "learning_rate": 0.0002,
+      "loss": 1.4678,
+      "step": 3930
+    },
+    {
+      "epoch": 3.838285435947394,
+      "grad_norm": 0.7049482464790344,
+      "learning_rate": 0.0002,
+      "loss": 1.4891,
+      "step": 3940
+    },
+    {
+      "epoch": 3.8480272771553823,
+      "grad_norm": 0.6533724665641785,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 3950
+    },
+    {
+      "epoch": 3.857769118363371,
+      "grad_norm": 0.7312499284744263,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 3960
+    },
+    {
+      "epoch": 3.867510959571359,
+      "grad_norm": 0.6858801245689392,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3970
+    },
+    {
+      "epoch": 3.877252800779347,
+      "grad_norm": 0.770423173904419,
+      "learning_rate": 0.0002,
+      "loss": 1.4423,
+      "step": 3980
+    },
+    {
+      "epoch": 3.886994641987336,
+      "grad_norm": 0.6987539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 3990
+    },
+    {
+      "epoch": 3.896736483195324,
+      "grad_norm": 0.7072722315788269,
+      "learning_rate": 0.0002,
+      "loss": 1.4791,
+      "step": 4000
+    },
+    {
+      "epoch": 3.906478324403312,
+      "grad_norm": 0.6492931842803955,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4010
+    },
+    {
+      "epoch": 3.9162201656113007,
+      "grad_norm": 0.7716232538223267,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 4020
+    },
+    {
+      "epoch": 3.925962006819289,
+      "grad_norm": 0.722949743270874,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 4030
+    },
+    {
+      "epoch": 3.935703848027277,
+      "grad_norm": 0.7434365749359131,
+      "learning_rate": 0.0002,
+      "loss": 1.3914,
+      "step": 4040
+    },
+    {
+      "epoch": 3.9454456892352656,
+      "grad_norm": 0.6691509485244751,
+      "learning_rate": 0.0002,
+      "loss": 1.4763,
+      "step": 4050
+    },
+    {
+      "epoch": 3.9551875304432538,
+      "grad_norm": 0.6850284337997437,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 4060
+    },
+    {
+      "epoch": 3.964929371651242,
+      "grad_norm": 0.6954452991485596,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4070
+    },
+    {
+      "epoch": 3.9746712128592305,
+      "grad_norm": 0.9316364526748657,
+      "learning_rate": 0.0002,
+      "loss": 1.417,
+      "step": 4080
+    },
+    {
+      "epoch": 3.9844130540672187,
+      "grad_norm": 0.6908289194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 4090
+    },
+    {
+      "epoch": 3.994154895275207,
+      "grad_norm": 0.666782021522522,
+      "learning_rate": 0.0002,
+      "loss": 1.4404,
+      "step": 4100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9233275651931763,
+      "eval_runtime": 55.9536,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 1.144,
+      "step": 4106
+    },
+    {
+      "epoch": 4.003896736483195,
+      "grad_norm": 0.7726166248321533,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 4110
+    },
+    {
+      "epoch": 4.013638577691184,
+      "grad_norm": 1.1338967084884644,
+      "learning_rate": 0.0002,
+      "loss": 1.1415,
+      "step": 4120
+    },
+    {
+      "epoch": 4.023380418899172,
+      "grad_norm": 0.9530029296875,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 4130
+    },
+    {
+      "epoch": 4.03312226010716,
+      "grad_norm": 1.1058554649353027,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 4140
+    },
+    {
+      "epoch": 4.042864101315149,
+      "grad_norm": 0.8765049576759338,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 4150
+    },
+    {
+      "epoch": 4.052605942523137,
+      "grad_norm": 1.1774667501449585,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4160
+    },
+    {
+      "epoch": 4.062347783731125,
+      "grad_norm": 0.9301433563232422,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 4170
+    },
+    {
+      "epoch": 4.072089624939114,
+      "grad_norm": 1.0196778774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.1807,
+      "step": 4180
+    },
+    {
+      "epoch": 4.081831466147102,
+      "grad_norm": 1.1380577087402344,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 4190
+    },
+    {
+      "epoch": 4.09157330735509,
+      "grad_norm": 0.9121319651603699,
+      "learning_rate": 0.0002,
+      "loss": 1.2521,
+      "step": 4200
+    },
+    {
+      "epoch": 4.101315148563079,
+      "grad_norm": 0.9495378732681274,
+      "learning_rate": 0.0002,
+      "loss": 1.1747,
+      "step": 4210
+    },
+    {
+      "epoch": 4.1110569897710665,
+      "grad_norm": 0.8058680295944214,
+      "learning_rate": 0.0002,
+      "loss": 1.1829,
+      "step": 4220
+    },
+    {
+      "epoch": 4.120798830979055,
+      "grad_norm": 1.000887393951416,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 4230
+    },
+    {
+      "epoch": 4.130540672187044,
+      "grad_norm": 0.9529102444648743,
+      "learning_rate": 0.0002,
+      "loss": 1.1947,
+      "step": 4240
+    },
+    {
+      "epoch": 4.140282513395031,
+      "grad_norm": 1.0257115364074707,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 4250
+    },
+    {
+      "epoch": 4.15002435460302,
+      "grad_norm": 0.9590303897857666,
+      "learning_rate": 0.0002,
+      "loss": 1.2293,
+      "step": 4260
+    },
+    {
+      "epoch": 4.159766195811009,
+      "grad_norm": 1.065291166305542,
+      "learning_rate": 0.0002,
+      "loss": 1.1918,
+      "step": 4270
+    },
+    {
+      "epoch": 4.169508037018996,
+      "grad_norm": 0.8819697499275208,
+      "learning_rate": 0.0002,
+      "loss": 1.2323,
+      "step": 4280
+    },
+    {
+      "epoch": 4.179249878226985,
+      "grad_norm": 1.0335261821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 4290
+    },
+    {
+      "epoch": 4.1889917194349735,
+      "grad_norm": 0.8872809410095215,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 4300
+    },
+    {
+      "epoch": 4.198733560642961,
+      "grad_norm": 0.9883159399032593,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 4310
+    },
+    {
+      "epoch": 4.20847540185095,
+      "grad_norm": 1.0254192352294922,
+      "learning_rate": 0.0002,
+      "loss": 1.2544,
+      "step": 4320
+    },
+    {
+      "epoch": 4.218217243058938,
+      "grad_norm": 0.9432600736618042,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 4330
+    },
+    {
+      "epoch": 4.227959084266926,
+      "grad_norm": 1.1008676290512085,
+      "learning_rate": 0.0002,
+      "loss": 1.2684,
+      "step": 4340
+    },
+    {
+      "epoch": 4.237700925474915,
+      "grad_norm": 1.0829699039459229,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 4350
+    },
+    {
+      "epoch": 4.247442766682903,
+      "grad_norm": 1.016847848892212,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4360
+    },
+    {
+      "epoch": 4.257184607890891,
+      "grad_norm": 0.8924864530563354,
+      "learning_rate": 0.0002,
+      "loss": 1.2375,
+      "step": 4370
+    },
+    {
+      "epoch": 4.26692644909888,
+      "grad_norm": 0.9300530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.1987,
+      "step": 4380
+    },
+    {
+      "epoch": 4.276668290306868,
+      "grad_norm": 0.9684814810752869,
+      "learning_rate": 0.0002,
+      "loss": 1.1696,
+      "step": 4390
+    },
+    {
+      "epoch": 4.286410131514856,
+      "grad_norm": 0.9916250705718994,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 4400
+    },
+    {
+      "epoch": 4.2961519727228445,
+      "grad_norm": 0.903680145740509,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 4410
+    },
+    {
+      "epoch": 4.305893813930833,
+      "grad_norm": 0.8713505268096924,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 4420
+    },
+    {
+      "epoch": 4.315635655138821,
+      "grad_norm": 0.9983905553817749,
+      "learning_rate": 0.0002,
+      "loss": 1.1957,
+      "step": 4430
+    },
+    {
+      "epoch": 4.3253774963468095,
+      "grad_norm": 1.1689040660858154,
+      "learning_rate": 0.0002,
+      "loss": 1.2676,
+      "step": 4440
+    },
+    {
+      "epoch": 4.335119337554798,
+      "grad_norm": 0.9316853880882263,
+      "learning_rate": 0.0002,
+      "loss": 1.2166,
+      "step": 4450
+    },
+    {
+      "epoch": 4.344861178762786,
+      "grad_norm": 0.9175887107849121,
+      "learning_rate": 0.0002,
+      "loss": 1.222,
+      "step": 4460
+    },
+    {
+      "epoch": 4.354603019970774,
+      "grad_norm": 0.9348906874656677,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4470
+    },
+    {
+      "epoch": 4.364344861178763,
+      "grad_norm": 0.9727016687393188,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 4480
+    },
+    {
+      "epoch": 4.374086702386751,
+      "grad_norm": 0.9843429923057556,
+      "learning_rate": 0.0002,
+      "loss": 1.2616,
+      "step": 4490
+    },
+    {
+      "epoch": 4.383828543594739,
+      "grad_norm": 0.9615852236747742,
+      "learning_rate": 0.0002,
+      "loss": 1.2488,
+      "step": 4500
+    },
+    {
+      "epoch": 4.393570384802728,
+      "grad_norm": 0.9688583612442017,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 4510
+    },
+    {
+      "epoch": 4.403312226010716,
+      "grad_norm": 0.9933668375015259,
+      "learning_rate": 0.0002,
+      "loss": 1.2546,
+      "step": 4520
+    },
+    {
+      "epoch": 4.413054067218704,
+      "grad_norm": 1.0626686811447144,
+      "learning_rate": 0.0002,
+      "loss": 1.2355,
+      "step": 4530
+    },
+    {
+      "epoch": 4.422795908426693,
+      "grad_norm": 0.9536267518997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 4540
+    },
+    {
+      "epoch": 4.432537749634681,
+      "grad_norm": 0.9777140021324158,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4550
+    },
+    {
+      "epoch": 4.442279590842669,
+      "grad_norm": 0.980780839920044,
+      "learning_rate": 0.0002,
+      "loss": 1.2878,
+      "step": 4560
+    },
+    {
+      "epoch": 4.452021432050658,
+      "grad_norm": 1.0147196054458618,
+      "learning_rate": 0.0002,
+      "loss": 1.2597,
+      "step": 4570
+    },
+    {
+      "epoch": 4.461763273258645,
+      "grad_norm": 0.9763361811637878,
+      "learning_rate": 0.0002,
+      "loss": 1.2148,
+      "step": 4580
+    },
+    {
+      "epoch": 4.471505114466634,
+      "grad_norm": 1.0300798416137695,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 4590
+    },
+    {
+      "epoch": 4.481246955674623,
+      "grad_norm": 0.8833121657371521,
+      "learning_rate": 0.0002,
+      "loss": 1.2665,
+      "step": 4600
+    },
+    {
+      "epoch": 4.490988796882611,
+      "grad_norm": 1.1214020252227783,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 4610
+    },
+    {
+      "epoch": 4.500730638090599,
+      "grad_norm": 0.8843787908554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2579,
+      "step": 4620
+    },
+    {
+      "epoch": 4.5104724792985875,
+      "grad_norm": 0.9942020773887634,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 4630
+    },
+    {
+      "epoch": 4.520214320506576,
+      "grad_norm": 1.0033202171325684,
+      "learning_rate": 0.0002,
+      "loss": 1.3172,
+      "step": 4640
+    },
+    {
+      "epoch": 4.529956161714564,
+      "grad_norm": 0.8767235279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2024,
+      "step": 4650
+    },
+    {
+      "epoch": 4.539698002922552,
+      "grad_norm": 1.0117276906967163,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 4660
+    },
+    {
+      "epoch": 4.549439844130541,
+      "grad_norm": 1.2787362337112427,
+      "learning_rate": 0.0002,
+      "loss": 1.2911,
+      "step": 4670
+    },
+    {
+      "epoch": 4.559181685338529,
+      "grad_norm": 0.8824878931045532,
+      "learning_rate": 0.0002,
+      "loss": 1.2603,
+      "step": 4680
+    },
+    {
+      "epoch": 4.568923526546517,
+      "grad_norm": 0.9209560751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4690
+    },
+    {
+      "epoch": 4.578665367754506,
+      "grad_norm": 1.1064010858535767,
+      "learning_rate": 0.0002,
+      "loss": 1.1916,
+      "step": 4700
+    },
+    {
+      "epoch": 4.588407208962494,
+      "grad_norm": 0.8914572596549988,
+      "learning_rate": 0.0002,
+      "loss": 1.2217,
+      "step": 4710
+    },
+    {
+      "epoch": 4.598149050170482,
+      "grad_norm": 1.0412265062332153,
+      "learning_rate": 0.0002,
+      "loss": 1.2861,
+      "step": 4720
+    },
+    {
+      "epoch": 4.607890891378471,
+      "grad_norm": 1.1950221061706543,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4730
+    },
+    {
+      "epoch": 4.617632732586459,
+      "grad_norm": 0.8938062787055969,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 4740
+    },
+    {
+      "epoch": 4.627374573794447,
+      "grad_norm": 0.9849569201469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4750
+    },
+    {
+      "epoch": 4.637116415002436,
+      "grad_norm": 1.0081515312194824,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 4760
+    },
+    {
+      "epoch": 4.6468582562104235,
+      "grad_norm": 0.8566309213638306,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 4770
+    },
+    {
+      "epoch": 4.656600097418412,
+      "grad_norm": 1.1750118732452393,
+      "learning_rate": 0.0002,
+      "loss": 1.2723,
+      "step": 4780
+    },
+    {
+      "epoch": 4.666341938626401,
+      "grad_norm": 0.925502598285675,
+      "learning_rate": 0.0002,
+      "loss": 1.2537,
+      "step": 4790
+    },
+    {
+      "epoch": 4.676083779834388,
+      "grad_norm": 1.0402472019195557,
+      "learning_rate": 0.0002,
+      "loss": 1.2146,
+      "step": 4800
+    },
+    {
+      "epoch": 4.685825621042377,
+      "grad_norm": 0.9772472977638245,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 4810
+    },
+    {
+      "epoch": 4.695567462250366,
+      "grad_norm": 0.9082779288291931,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 4820
+    },
+    {
+      "epoch": 4.705309303458353,
+      "grad_norm": 0.8026862740516663,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 4830
+    },
+    {
+      "epoch": 4.715051144666342,
+      "grad_norm": 1.1631089448928833,
+      "learning_rate": 0.0002,
+      "loss": 1.3369,
+      "step": 4840
+    },
+    {
+      "epoch": 4.7247929858743305,
+      "grad_norm": 0.9384787678718567,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 4850
+    },
+    {
+      "epoch": 4.734534827082318,
+      "grad_norm": 1.2151581048965454,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 4860
+    },
+    {
+      "epoch": 4.744276668290307,
+      "grad_norm": 0.9679436087608337,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 4870
+    },
+    {
+      "epoch": 4.754018509498295,
+      "grad_norm": 0.8352158069610596,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 4880
+    },
+    {
+      "epoch": 4.763760350706283,
+      "grad_norm": 1.0205804109573364,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 4890
+    },
+    {
+      "epoch": 4.773502191914272,
+      "grad_norm": 0.9814772605895996,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 4900
+    },
+    {
+      "epoch": 4.78324403312226,
+      "grad_norm": 1.002854347229004,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 4910
+    },
+    {
+      "epoch": 4.792985874330248,
+      "grad_norm": 1.1609505414962769,
+      "learning_rate": 0.0002,
+      "loss": 1.3143,
+      "step": 4920
+    },
+    {
+      "epoch": 4.802727715538237,
+      "grad_norm": 0.9354982376098633,
+      "learning_rate": 0.0002,
+      "loss": 1.3166,
+      "step": 4930
+    },
+    {
+      "epoch": 4.812469556746225,
+      "grad_norm": 0.9761685729026794,
+      "learning_rate": 0.0002,
+      "loss": 1.2978,
+      "step": 4940
+    },
+    {
+      "epoch": 4.822211397954213,
+      "grad_norm": 1.0604596138000488,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 4950
+    },
+    {
+      "epoch": 4.8319532391622015,
+      "grad_norm": 1.0902808904647827,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 4960
+    },
+    {
+      "epoch": 4.84169508037019,
+      "grad_norm": 1.0174955129623413,
+      "learning_rate": 0.0002,
+      "loss": 1.3073,
+      "step": 4970
+    },
+    {
+      "epoch": 4.851436921578179,
+      "grad_norm": 1.0995253324508667,
+      "learning_rate": 0.0002,
+      "loss": 1.3141,
+      "step": 4980
+    },
+    {
+      "epoch": 4.8611787627861665,
+      "grad_norm": 0.880993127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4990
+    },
+    {
+      "epoch": 4.870920603994155,
+      "grad_norm": 0.9472237825393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2547,
+      "step": 5000
+    },
+    {
+      "epoch": 4.880662445202143,
+      "grad_norm": 0.9504236578941345,
+      "learning_rate": 0.0002,
+      "loss": 1.4078,
+      "step": 5010
+    },
+    {
+      "epoch": 4.890404286410131,
+      "grad_norm": 1.1261742115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2791,
+      "step": 5020
+    },
+    {
+      "epoch": 4.90014612761812,
+      "grad_norm": 0.904674768447876,
+      "learning_rate": 0.0002,
+      "loss": 1.3707,
+      "step": 5030
+    },
+    {
+      "epoch": 4.909887968826109,
+      "grad_norm": 0.8828991055488586,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 5040
+    },
+    {
+      "epoch": 4.919629810034096,
+      "grad_norm": 1.0156532526016235,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 5050
+    },
+    {
+      "epoch": 4.929371651242085,
+      "grad_norm": 0.8975168466567993,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 5060
+    },
+    {
+      "epoch": 4.939113492450073,
+      "grad_norm": 0.9787213802337646,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 5070
+    },
+    {
+      "epoch": 4.948855333658061,
+      "grad_norm": 1.0801568031311035,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 5080
+    },
+    {
+      "epoch": 4.95859717486605,
+      "grad_norm": 1.0655089616775513,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 5090
+    },
+    {
+      "epoch": 4.968339016074038,
+      "grad_norm": 0.8941320180892944,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 5100
+    },
+    {
+      "epoch": 4.978080857282026,
+      "grad_norm": 1.050621747970581,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 5110
+    },
+    {
+      "epoch": 4.987822698490015,
+      "grad_norm": 0.9724781513214111,
+      "learning_rate": 0.0002,
+      "loss": 1.3791,
+      "step": 5120
+    },
+    {
+      "epoch": 4.997564539698003,
+      "grad_norm": 0.9850538969039917,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 5130
+    },
+    {
+      "epoch": 4.9995129079396,
+      "eval_loss": 2.0824170112609863,
+      "eval_runtime": 55.592,
+      "eval_samples_per_second": 9.12,
+      "eval_steps_per_second": 1.151,
+      "step": 5132
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 2.636244244365312e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-5132/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f6017ad2003beb2b8b1e6489151fda4210695ba6
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f05a8db3be673b644ec72212b7426226dfc6a4143b9db1d52d06b769defd957e
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..aeb23b37299e91905445e7a634bc47b74af209eb
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc1bf22316ad6f9c423bce00fc88b52faa240878bb818509a7d6bd35e32a04af
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a26dad7804e1958688284db43885b5075b1f773b
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3bc5a941d2094fad4f0f385bae3ab22f53e7e0fc54bf4e91b5db83b189760ee
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f64ade66f62200ea3b67d81f19c9b595c8d0b98c
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e10b88e0762b2dc9ec6a69bbd0025951b3a06dcda517a323ae83fcf1ba4038e
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..dc395a0a8299d1a55b5b2397714b16d610cd9cc9
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/trainer_state.json
@@ -0,0 +1,4386 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 6159,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    },
+    {
+      "epoch": 3.0004870920603994,
+      "grad_norm": 0.49992576241493225,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 3.0102289332683876,
+      "grad_norm": 0.8242783546447754,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3090
+    },
+    {
+      "epoch": 3.019970774476376,
+      "grad_norm": 0.6330569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3100
+    },
+    {
+      "epoch": 3.0297126156843643,
+      "grad_norm": 0.566097617149353,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 3110
+    },
+    {
+      "epoch": 3.0394544568923525,
+      "grad_norm": 0.6337586045265198,
+      "learning_rate": 0.0002,
+      "loss": 1.4365,
+      "step": 3120
+    },
+    {
+      "epoch": 3.049196298100341,
+      "grad_norm": 0.7339403033256531,
+      "learning_rate": 0.0002,
+      "loss": 1.3916,
+      "step": 3130
+    },
+    {
+      "epoch": 3.0589381393083293,
+      "grad_norm": 0.7187346816062927,
+      "learning_rate": 0.0002,
+      "loss": 1.4617,
+      "step": 3140
+    },
+    {
+      "epoch": 3.0686799805163174,
+      "grad_norm": 0.7116255760192871,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 3150
+    },
+    {
+      "epoch": 3.078421821724306,
+      "grad_norm": 0.6493807435035706,
+      "learning_rate": 0.0002,
+      "loss": 1.4452,
+      "step": 3160
+    },
+    {
+      "epoch": 3.088163662932294,
+      "grad_norm": 0.6777266263961792,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 3170
+    },
+    {
+      "epoch": 3.0979055041402823,
+      "grad_norm": 0.6342006325721741,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 3180
+    },
+    {
+      "epoch": 3.107647345348271,
+      "grad_norm": 0.6608964204788208,
+      "learning_rate": 0.0002,
+      "loss": 1.4748,
+      "step": 3190
+    },
+    {
+      "epoch": 3.117389186556259,
+      "grad_norm": 0.7230247259140015,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 3200
+    },
+    {
+      "epoch": 3.1271310277642472,
+      "grad_norm": 0.650368332862854,
+      "learning_rate": 0.0002,
+      "loss": 1.4049,
+      "step": 3210
+    },
+    {
+      "epoch": 3.136872868972236,
+      "grad_norm": 0.7319342494010925,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 3220
+    },
+    {
+      "epoch": 3.146614710180224,
+      "grad_norm": 0.7159963846206665,
+      "learning_rate": 0.0002,
+      "loss": 1.3872,
+      "step": 3230
+    },
+    {
+      "epoch": 3.156356551388212,
+      "grad_norm": 0.8905230164527893,
+      "learning_rate": 0.0002,
+      "loss": 1.5076,
+      "step": 3240
+    },
+    {
+      "epoch": 3.1660983925962007,
+      "grad_norm": 0.6920804381370544,
+      "learning_rate": 0.0002,
+      "loss": 1.3161,
+      "step": 3250
+    },
+    {
+      "epoch": 3.175840233804189,
+      "grad_norm": 0.6782063841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.3786,
+      "step": 3260
+    },
+    {
+      "epoch": 3.1855820750121775,
+      "grad_norm": 0.735325276851654,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 3270
+    },
+    {
+      "epoch": 3.1953239162201656,
+      "grad_norm": 0.6657978296279907,
+      "learning_rate": 0.0002,
+      "loss": 1.4027,
+      "step": 3280
+    },
+    {
+      "epoch": 3.205065757428154,
+      "grad_norm": 0.771315336227417,
+      "learning_rate": 0.0002,
+      "loss": 1.3456,
+      "step": 3290
+    },
+    {
+      "epoch": 3.2148075986361424,
+      "grad_norm": 0.6492983102798462,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 3300
+    },
+    {
+      "epoch": 3.2245494398441306,
+      "grad_norm": 0.7513770461082458,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3310
+    },
+    {
+      "epoch": 3.2342912810521187,
+      "grad_norm": 0.7091423869132996,
+      "learning_rate": 0.0002,
+      "loss": 1.4032,
+      "step": 3320
+    },
+    {
+      "epoch": 3.2440331222601073,
+      "grad_norm": 0.6663975119590759,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 3330
+    },
+    {
+      "epoch": 3.2537749634680955,
+      "grad_norm": 0.6813122034072876,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 3340
+    },
+    {
+      "epoch": 3.2635168046760836,
+      "grad_norm": 0.6602569818496704,
+      "learning_rate": 0.0002,
+      "loss": 1.3681,
+      "step": 3350
+    },
+    {
+      "epoch": 3.2732586458840722,
+      "grad_norm": 0.718270480632782,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 3360
+    },
+    {
+      "epoch": 3.2830004870920604,
+      "grad_norm": 0.6884173154830933,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 3370
+    },
+    {
+      "epoch": 3.2927423283000485,
+      "grad_norm": 0.7039775848388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3380
+    },
+    {
+      "epoch": 3.302484169508037,
+      "grad_norm": 0.7444299459457397,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3390
+    },
+    {
+      "epoch": 3.3122260107160253,
+      "grad_norm": 0.7187064290046692,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 3400
+    },
+    {
+      "epoch": 3.3219678519240134,
+      "grad_norm": 0.599396288394928,
+      "learning_rate": 0.0002,
+      "loss": 1.3684,
+      "step": 3410
+    },
+    {
+      "epoch": 3.331709693132002,
+      "grad_norm": 0.7670390009880066,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3420
+    },
+    {
+      "epoch": 3.34145153433999,
+      "grad_norm": 0.6654478311538696,
+      "learning_rate": 0.0002,
+      "loss": 1.4411,
+      "step": 3430
+    },
+    {
+      "epoch": 3.351193375547979,
+      "grad_norm": 0.6644385457038879,
+      "learning_rate": 0.0002,
+      "loss": 1.4257,
+      "step": 3440
+    },
+    {
+      "epoch": 3.360935216755967,
+      "grad_norm": 0.6974098086357117,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3450
+    },
+    {
+      "epoch": 3.370677057963955,
+      "grad_norm": 0.7350399494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 3460
+    },
+    {
+      "epoch": 3.3804188991719437,
+      "grad_norm": 0.714721143245697,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 3470
+    },
+    {
+      "epoch": 3.390160740379932,
+      "grad_norm": 0.7006027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 3480
+    },
+    {
+      "epoch": 3.39990258158792,
+      "grad_norm": 0.6767925024032593,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 3490
+    },
+    {
+      "epoch": 3.4096444227959086,
+      "grad_norm": 0.6721355319023132,
+      "learning_rate": 0.0002,
+      "loss": 1.4116,
+      "step": 3500
+    },
+    {
+      "epoch": 3.419386264003897,
+      "grad_norm": 0.6845725178718567,
+      "learning_rate": 0.0002,
+      "loss": 1.443,
+      "step": 3510
+    },
+    {
+      "epoch": 3.429128105211885,
+      "grad_norm": 0.6882196664810181,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 3520
+    },
+    {
+      "epoch": 3.4388699464198735,
+      "grad_norm": 0.7663240432739258,
+      "learning_rate": 0.0002,
+      "loss": 1.4962,
+      "step": 3530
+    },
+    {
+      "epoch": 3.4486117876278617,
+      "grad_norm": 0.6304219365119934,
+      "learning_rate": 0.0002,
+      "loss": 1.4644,
+      "step": 3540
+    },
+    {
+      "epoch": 3.45835362883585,
+      "grad_norm": 0.668678879737854,
+      "learning_rate": 0.0002,
+      "loss": 1.4918,
+      "step": 3550
+    },
+    {
+      "epoch": 3.4680954700438384,
+      "grad_norm": 0.7526912093162537,
+      "learning_rate": 0.0002,
+      "loss": 1.4874,
+      "step": 3560
+    },
+    {
+      "epoch": 3.4778373112518266,
+      "grad_norm": 1.089495301246643,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 3570
+    },
+    {
+      "epoch": 3.4875791524598148,
+      "grad_norm": 0.7282902002334595,
+      "learning_rate": 0.0002,
+      "loss": 1.3871,
+      "step": 3580
+    },
+    {
+      "epoch": 3.4973209936678034,
+      "grad_norm": 0.6540156602859497,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3590
+    },
+    {
+      "epoch": 3.5070628348757915,
+      "grad_norm": 0.6449568867683411,
+      "learning_rate": 0.0002,
+      "loss": 1.4367,
+      "step": 3600
+    },
+    {
+      "epoch": 3.5168046760837797,
+      "grad_norm": 0.7262216210365295,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 3610
+    },
+    {
+      "epoch": 3.5265465172917683,
+      "grad_norm": 0.6048615574836731,
+      "learning_rate": 0.0002,
+      "loss": 1.4374,
+      "step": 3620
+    },
+    {
+      "epoch": 3.5362883584997564,
+      "grad_norm": 0.6780537366867065,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 3630
+    },
+    {
+      "epoch": 3.5460301997077446,
+      "grad_norm": 0.6851925253868103,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 3640
+    },
+    {
+      "epoch": 3.555772040915733,
+      "grad_norm": 0.6530634164810181,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 3650
+    },
+    {
+      "epoch": 3.5655138821237213,
+      "grad_norm": 0.7193992733955383,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 3660
+    },
+    {
+      "epoch": 3.5752557233317095,
+      "grad_norm": 0.767496645450592,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 3670
+    },
+    {
+      "epoch": 3.584997564539698,
+      "grad_norm": 0.6912919282913208,
+      "learning_rate": 0.0002,
+      "loss": 1.4824,
+      "step": 3680
+    },
+    {
+      "epoch": 3.5947394057476862,
+      "grad_norm": 0.7383436560630798,
+      "learning_rate": 0.0002,
+      "loss": 1.4497,
+      "step": 3690
+    },
+    {
+      "epoch": 3.6044812469556744,
+      "grad_norm": 0.6746662855148315,
+      "learning_rate": 0.0002,
+      "loss": 1.4822,
+      "step": 3700
+    },
+    {
+      "epoch": 3.614223088163663,
+      "grad_norm": 0.6885138750076294,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 3710
+    },
+    {
+      "epoch": 3.623964929371651,
+      "grad_norm": 0.6694392561912537,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 3720
+    },
+    {
+      "epoch": 3.6337067705796393,
+      "grad_norm": 0.812358021736145,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 3730
+    },
+    {
+      "epoch": 3.643448611787628,
+      "grad_norm": 0.7267130017280579,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 3740
+    },
+    {
+      "epoch": 3.653190452995616,
+      "grad_norm": 0.6958749294281006,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 3750
+    },
+    {
+      "epoch": 3.6629322942036042,
+      "grad_norm": 0.6805673241615295,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3760
+    },
+    {
+      "epoch": 3.672674135411593,
+      "grad_norm": 0.7184410095214844,
+      "learning_rate": 0.0002,
+      "loss": 1.4338,
+      "step": 3770
+    },
+    {
+      "epoch": 3.682415976619581,
+      "grad_norm": 0.7716330289840698,
+      "learning_rate": 0.0002,
+      "loss": 1.3935,
+      "step": 3780
+    },
+    {
+      "epoch": 3.6921578178275696,
+      "grad_norm": 0.6675831079483032,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 3790
+    },
+    {
+      "epoch": 3.7018996590355577,
+      "grad_norm": 0.6480095386505127,
+      "learning_rate": 0.0002,
+      "loss": 1.401,
+      "step": 3800
+    },
+    {
+      "epoch": 3.711641500243546,
+      "grad_norm": 0.6559418439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 3810
+    },
+    {
+      "epoch": 3.7213833414515345,
+      "grad_norm": 0.6596545577049255,
+      "learning_rate": 0.0002,
+      "loss": 1.4341,
+      "step": 3820
+    },
+    {
+      "epoch": 3.7311251826595226,
+      "grad_norm": 0.7172950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3830
+    },
+    {
+      "epoch": 3.740867023867511,
+      "grad_norm": 0.796148419380188,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 3840
+    },
+    {
+      "epoch": 3.7506088650754994,
+      "grad_norm": 0.6600322723388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4992,
+      "step": 3850
+    },
+    {
+      "epoch": 3.7603507062834876,
+      "grad_norm": 0.6776387691497803,
+      "learning_rate": 0.0002,
+      "loss": 1.4201,
+      "step": 3860
+    },
+    {
+      "epoch": 3.770092547491476,
+      "grad_norm": 0.7768304347991943,
+      "learning_rate": 0.0002,
+      "loss": 1.3893,
+      "step": 3870
+    },
+    {
+      "epoch": 3.7798343886994643,
+      "grad_norm": 1.0579794645309448,
+      "learning_rate": 0.0002,
+      "loss": 1.4886,
+      "step": 3880
+    },
+    {
+      "epoch": 3.7895762299074525,
+      "grad_norm": 0.6757252812385559,
+      "learning_rate": 0.0002,
+      "loss": 1.4556,
+      "step": 3890
+    },
+    {
+      "epoch": 3.799318071115441,
+      "grad_norm": 0.6706996560096741,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 3900
+    },
+    {
+      "epoch": 3.809059912323429,
+      "grad_norm": 0.7026948928833008,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 3910
+    },
+    {
+      "epoch": 3.8188017535314174,
+      "grad_norm": 0.6437768340110779,
+      "learning_rate": 0.0002,
+      "loss": 1.5487,
+      "step": 3920
+    },
+    {
+      "epoch": 3.828543594739406,
+      "grad_norm": 0.7015706300735474,
+      "learning_rate": 0.0002,
+      "loss": 1.4678,
+      "step": 3930
+    },
+    {
+      "epoch": 3.838285435947394,
+      "grad_norm": 0.7049482464790344,
+      "learning_rate": 0.0002,
+      "loss": 1.4891,
+      "step": 3940
+    },
+    {
+      "epoch": 3.8480272771553823,
+      "grad_norm": 0.6533724665641785,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 3950
+    },
+    {
+      "epoch": 3.857769118363371,
+      "grad_norm": 0.7312499284744263,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 3960
+    },
+    {
+      "epoch": 3.867510959571359,
+      "grad_norm": 0.6858801245689392,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3970
+    },
+    {
+      "epoch": 3.877252800779347,
+      "grad_norm": 0.770423173904419,
+      "learning_rate": 0.0002,
+      "loss": 1.4423,
+      "step": 3980
+    },
+    {
+      "epoch": 3.886994641987336,
+      "grad_norm": 0.6987539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 3990
+    },
+    {
+      "epoch": 3.896736483195324,
+      "grad_norm": 0.7072722315788269,
+      "learning_rate": 0.0002,
+      "loss": 1.4791,
+      "step": 4000
+    },
+    {
+      "epoch": 3.906478324403312,
+      "grad_norm": 0.6492931842803955,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4010
+    },
+    {
+      "epoch": 3.9162201656113007,
+      "grad_norm": 0.7716232538223267,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 4020
+    },
+    {
+      "epoch": 3.925962006819289,
+      "grad_norm": 0.722949743270874,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 4030
+    },
+    {
+      "epoch": 3.935703848027277,
+      "grad_norm": 0.7434365749359131,
+      "learning_rate": 0.0002,
+      "loss": 1.3914,
+      "step": 4040
+    },
+    {
+      "epoch": 3.9454456892352656,
+      "grad_norm": 0.6691509485244751,
+      "learning_rate": 0.0002,
+      "loss": 1.4763,
+      "step": 4050
+    },
+    {
+      "epoch": 3.9551875304432538,
+      "grad_norm": 0.6850284337997437,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 4060
+    },
+    {
+      "epoch": 3.964929371651242,
+      "grad_norm": 0.6954452991485596,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4070
+    },
+    {
+      "epoch": 3.9746712128592305,
+      "grad_norm": 0.9316364526748657,
+      "learning_rate": 0.0002,
+      "loss": 1.417,
+      "step": 4080
+    },
+    {
+      "epoch": 3.9844130540672187,
+      "grad_norm": 0.6908289194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 4090
+    },
+    {
+      "epoch": 3.994154895275207,
+      "grad_norm": 0.666782021522522,
+      "learning_rate": 0.0002,
+      "loss": 1.4404,
+      "step": 4100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9233275651931763,
+      "eval_runtime": 55.9536,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 1.144,
+      "step": 4106
+    },
+    {
+      "epoch": 4.003896736483195,
+      "grad_norm": 0.7726166248321533,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 4110
+    },
+    {
+      "epoch": 4.013638577691184,
+      "grad_norm": 1.1338967084884644,
+      "learning_rate": 0.0002,
+      "loss": 1.1415,
+      "step": 4120
+    },
+    {
+      "epoch": 4.023380418899172,
+      "grad_norm": 0.9530029296875,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 4130
+    },
+    {
+      "epoch": 4.03312226010716,
+      "grad_norm": 1.1058554649353027,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 4140
+    },
+    {
+      "epoch": 4.042864101315149,
+      "grad_norm": 0.8765049576759338,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 4150
+    },
+    {
+      "epoch": 4.052605942523137,
+      "grad_norm": 1.1774667501449585,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4160
+    },
+    {
+      "epoch": 4.062347783731125,
+      "grad_norm": 0.9301433563232422,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 4170
+    },
+    {
+      "epoch": 4.072089624939114,
+      "grad_norm": 1.0196778774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.1807,
+      "step": 4180
+    },
+    {
+      "epoch": 4.081831466147102,
+      "grad_norm": 1.1380577087402344,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 4190
+    },
+    {
+      "epoch": 4.09157330735509,
+      "grad_norm": 0.9121319651603699,
+      "learning_rate": 0.0002,
+      "loss": 1.2521,
+      "step": 4200
+    },
+    {
+      "epoch": 4.101315148563079,
+      "grad_norm": 0.9495378732681274,
+      "learning_rate": 0.0002,
+      "loss": 1.1747,
+      "step": 4210
+    },
+    {
+      "epoch": 4.1110569897710665,
+      "grad_norm": 0.8058680295944214,
+      "learning_rate": 0.0002,
+      "loss": 1.1829,
+      "step": 4220
+    },
+    {
+      "epoch": 4.120798830979055,
+      "grad_norm": 1.000887393951416,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 4230
+    },
+    {
+      "epoch": 4.130540672187044,
+      "grad_norm": 0.9529102444648743,
+      "learning_rate": 0.0002,
+      "loss": 1.1947,
+      "step": 4240
+    },
+    {
+      "epoch": 4.140282513395031,
+      "grad_norm": 1.0257115364074707,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 4250
+    },
+    {
+      "epoch": 4.15002435460302,
+      "grad_norm": 0.9590303897857666,
+      "learning_rate": 0.0002,
+      "loss": 1.2293,
+      "step": 4260
+    },
+    {
+      "epoch": 4.159766195811009,
+      "grad_norm": 1.065291166305542,
+      "learning_rate": 0.0002,
+      "loss": 1.1918,
+      "step": 4270
+    },
+    {
+      "epoch": 4.169508037018996,
+      "grad_norm": 0.8819697499275208,
+      "learning_rate": 0.0002,
+      "loss": 1.2323,
+      "step": 4280
+    },
+    {
+      "epoch": 4.179249878226985,
+      "grad_norm": 1.0335261821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 4290
+    },
+    {
+      "epoch": 4.1889917194349735,
+      "grad_norm": 0.8872809410095215,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 4300
+    },
+    {
+      "epoch": 4.198733560642961,
+      "grad_norm": 0.9883159399032593,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 4310
+    },
+    {
+      "epoch": 4.20847540185095,
+      "grad_norm": 1.0254192352294922,
+      "learning_rate": 0.0002,
+      "loss": 1.2544,
+      "step": 4320
+    },
+    {
+      "epoch": 4.218217243058938,
+      "grad_norm": 0.9432600736618042,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 4330
+    },
+    {
+      "epoch": 4.227959084266926,
+      "grad_norm": 1.1008676290512085,
+      "learning_rate": 0.0002,
+      "loss": 1.2684,
+      "step": 4340
+    },
+    {
+      "epoch": 4.237700925474915,
+      "grad_norm": 1.0829699039459229,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 4350
+    },
+    {
+      "epoch": 4.247442766682903,
+      "grad_norm": 1.016847848892212,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4360
+    },
+    {
+      "epoch": 4.257184607890891,
+      "grad_norm": 0.8924864530563354,
+      "learning_rate": 0.0002,
+      "loss": 1.2375,
+      "step": 4370
+    },
+    {
+      "epoch": 4.26692644909888,
+      "grad_norm": 0.9300530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.1987,
+      "step": 4380
+    },
+    {
+      "epoch": 4.276668290306868,
+      "grad_norm": 0.9684814810752869,
+      "learning_rate": 0.0002,
+      "loss": 1.1696,
+      "step": 4390
+    },
+    {
+      "epoch": 4.286410131514856,
+      "grad_norm": 0.9916250705718994,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 4400
+    },
+    {
+      "epoch": 4.2961519727228445,
+      "grad_norm": 0.903680145740509,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 4410
+    },
+    {
+      "epoch": 4.305893813930833,
+      "grad_norm": 0.8713505268096924,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 4420
+    },
+    {
+      "epoch": 4.315635655138821,
+      "grad_norm": 0.9983905553817749,
+      "learning_rate": 0.0002,
+      "loss": 1.1957,
+      "step": 4430
+    },
+    {
+      "epoch": 4.3253774963468095,
+      "grad_norm": 1.1689040660858154,
+      "learning_rate": 0.0002,
+      "loss": 1.2676,
+      "step": 4440
+    },
+    {
+      "epoch": 4.335119337554798,
+      "grad_norm": 0.9316853880882263,
+      "learning_rate": 0.0002,
+      "loss": 1.2166,
+      "step": 4450
+    },
+    {
+      "epoch": 4.344861178762786,
+      "grad_norm": 0.9175887107849121,
+      "learning_rate": 0.0002,
+      "loss": 1.222,
+      "step": 4460
+    },
+    {
+      "epoch": 4.354603019970774,
+      "grad_norm": 0.9348906874656677,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4470
+    },
+    {
+      "epoch": 4.364344861178763,
+      "grad_norm": 0.9727016687393188,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 4480
+    },
+    {
+      "epoch": 4.374086702386751,
+      "grad_norm": 0.9843429923057556,
+      "learning_rate": 0.0002,
+      "loss": 1.2616,
+      "step": 4490
+    },
+    {
+      "epoch": 4.383828543594739,
+      "grad_norm": 0.9615852236747742,
+      "learning_rate": 0.0002,
+      "loss": 1.2488,
+      "step": 4500
+    },
+    {
+      "epoch": 4.393570384802728,
+      "grad_norm": 0.9688583612442017,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 4510
+    },
+    {
+      "epoch": 4.403312226010716,
+      "grad_norm": 0.9933668375015259,
+      "learning_rate": 0.0002,
+      "loss": 1.2546,
+      "step": 4520
+    },
+    {
+      "epoch": 4.413054067218704,
+      "grad_norm": 1.0626686811447144,
+      "learning_rate": 0.0002,
+      "loss": 1.2355,
+      "step": 4530
+    },
+    {
+      "epoch": 4.422795908426693,
+      "grad_norm": 0.9536267518997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 4540
+    },
+    {
+      "epoch": 4.432537749634681,
+      "grad_norm": 0.9777140021324158,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4550
+    },
+    {
+      "epoch": 4.442279590842669,
+      "grad_norm": 0.980780839920044,
+      "learning_rate": 0.0002,
+      "loss": 1.2878,
+      "step": 4560
+    },
+    {
+      "epoch": 4.452021432050658,
+      "grad_norm": 1.0147196054458618,
+      "learning_rate": 0.0002,
+      "loss": 1.2597,
+      "step": 4570
+    },
+    {
+      "epoch": 4.461763273258645,
+      "grad_norm": 0.9763361811637878,
+      "learning_rate": 0.0002,
+      "loss": 1.2148,
+      "step": 4580
+    },
+    {
+      "epoch": 4.471505114466634,
+      "grad_norm": 1.0300798416137695,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 4590
+    },
+    {
+      "epoch": 4.481246955674623,
+      "grad_norm": 0.8833121657371521,
+      "learning_rate": 0.0002,
+      "loss": 1.2665,
+      "step": 4600
+    },
+    {
+      "epoch": 4.490988796882611,
+      "grad_norm": 1.1214020252227783,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 4610
+    },
+    {
+      "epoch": 4.500730638090599,
+      "grad_norm": 0.8843787908554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2579,
+      "step": 4620
+    },
+    {
+      "epoch": 4.5104724792985875,
+      "grad_norm": 0.9942020773887634,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 4630
+    },
+    {
+      "epoch": 4.520214320506576,
+      "grad_norm": 1.0033202171325684,
+      "learning_rate": 0.0002,
+      "loss": 1.3172,
+      "step": 4640
+    },
+    {
+      "epoch": 4.529956161714564,
+      "grad_norm": 0.8767235279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2024,
+      "step": 4650
+    },
+    {
+      "epoch": 4.539698002922552,
+      "grad_norm": 1.0117276906967163,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 4660
+    },
+    {
+      "epoch": 4.549439844130541,
+      "grad_norm": 1.2787362337112427,
+      "learning_rate": 0.0002,
+      "loss": 1.2911,
+      "step": 4670
+    },
+    {
+      "epoch": 4.559181685338529,
+      "grad_norm": 0.8824878931045532,
+      "learning_rate": 0.0002,
+      "loss": 1.2603,
+      "step": 4680
+    },
+    {
+      "epoch": 4.568923526546517,
+      "grad_norm": 0.9209560751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4690
+    },
+    {
+      "epoch": 4.578665367754506,
+      "grad_norm": 1.1064010858535767,
+      "learning_rate": 0.0002,
+      "loss": 1.1916,
+      "step": 4700
+    },
+    {
+      "epoch": 4.588407208962494,
+      "grad_norm": 0.8914572596549988,
+      "learning_rate": 0.0002,
+      "loss": 1.2217,
+      "step": 4710
+    },
+    {
+      "epoch": 4.598149050170482,
+      "grad_norm": 1.0412265062332153,
+      "learning_rate": 0.0002,
+      "loss": 1.2861,
+      "step": 4720
+    },
+    {
+      "epoch": 4.607890891378471,
+      "grad_norm": 1.1950221061706543,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4730
+    },
+    {
+      "epoch": 4.617632732586459,
+      "grad_norm": 0.8938062787055969,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 4740
+    },
+    {
+      "epoch": 4.627374573794447,
+      "grad_norm": 0.9849569201469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4750
+    },
+    {
+      "epoch": 4.637116415002436,
+      "grad_norm": 1.0081515312194824,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 4760
+    },
+    {
+      "epoch": 4.6468582562104235,
+      "grad_norm": 0.8566309213638306,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 4770
+    },
+    {
+      "epoch": 4.656600097418412,
+      "grad_norm": 1.1750118732452393,
+      "learning_rate": 0.0002,
+      "loss": 1.2723,
+      "step": 4780
+    },
+    {
+      "epoch": 4.666341938626401,
+      "grad_norm": 0.925502598285675,
+      "learning_rate": 0.0002,
+      "loss": 1.2537,
+      "step": 4790
+    },
+    {
+      "epoch": 4.676083779834388,
+      "grad_norm": 1.0402472019195557,
+      "learning_rate": 0.0002,
+      "loss": 1.2146,
+      "step": 4800
+    },
+    {
+      "epoch": 4.685825621042377,
+      "grad_norm": 0.9772472977638245,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 4810
+    },
+    {
+      "epoch": 4.695567462250366,
+      "grad_norm": 0.9082779288291931,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 4820
+    },
+    {
+      "epoch": 4.705309303458353,
+      "grad_norm": 0.8026862740516663,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 4830
+    },
+    {
+      "epoch": 4.715051144666342,
+      "grad_norm": 1.1631089448928833,
+      "learning_rate": 0.0002,
+      "loss": 1.3369,
+      "step": 4840
+    },
+    {
+      "epoch": 4.7247929858743305,
+      "grad_norm": 0.9384787678718567,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 4850
+    },
+    {
+      "epoch": 4.734534827082318,
+      "grad_norm": 1.2151581048965454,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 4860
+    },
+    {
+      "epoch": 4.744276668290307,
+      "grad_norm": 0.9679436087608337,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 4870
+    },
+    {
+      "epoch": 4.754018509498295,
+      "grad_norm": 0.8352158069610596,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 4880
+    },
+    {
+      "epoch": 4.763760350706283,
+      "grad_norm": 1.0205804109573364,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 4890
+    },
+    {
+      "epoch": 4.773502191914272,
+      "grad_norm": 0.9814772605895996,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 4900
+    },
+    {
+      "epoch": 4.78324403312226,
+      "grad_norm": 1.002854347229004,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 4910
+    },
+    {
+      "epoch": 4.792985874330248,
+      "grad_norm": 1.1609505414962769,
+      "learning_rate": 0.0002,
+      "loss": 1.3143,
+      "step": 4920
+    },
+    {
+      "epoch": 4.802727715538237,
+      "grad_norm": 0.9354982376098633,
+      "learning_rate": 0.0002,
+      "loss": 1.3166,
+      "step": 4930
+    },
+    {
+      "epoch": 4.812469556746225,
+      "grad_norm": 0.9761685729026794,
+      "learning_rate": 0.0002,
+      "loss": 1.2978,
+      "step": 4940
+    },
+    {
+      "epoch": 4.822211397954213,
+      "grad_norm": 1.0604596138000488,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 4950
+    },
+    {
+      "epoch": 4.8319532391622015,
+      "grad_norm": 1.0902808904647827,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 4960
+    },
+    {
+      "epoch": 4.84169508037019,
+      "grad_norm": 1.0174955129623413,
+      "learning_rate": 0.0002,
+      "loss": 1.3073,
+      "step": 4970
+    },
+    {
+      "epoch": 4.851436921578179,
+      "grad_norm": 1.0995253324508667,
+      "learning_rate": 0.0002,
+      "loss": 1.3141,
+      "step": 4980
+    },
+    {
+      "epoch": 4.8611787627861665,
+      "grad_norm": 0.880993127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4990
+    },
+    {
+      "epoch": 4.870920603994155,
+      "grad_norm": 0.9472237825393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2547,
+      "step": 5000
+    },
+    {
+      "epoch": 4.880662445202143,
+      "grad_norm": 0.9504236578941345,
+      "learning_rate": 0.0002,
+      "loss": 1.4078,
+      "step": 5010
+    },
+    {
+      "epoch": 4.890404286410131,
+      "grad_norm": 1.1261742115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2791,
+      "step": 5020
+    },
+    {
+      "epoch": 4.90014612761812,
+      "grad_norm": 0.904674768447876,
+      "learning_rate": 0.0002,
+      "loss": 1.3707,
+      "step": 5030
+    },
+    {
+      "epoch": 4.909887968826109,
+      "grad_norm": 0.8828991055488586,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 5040
+    },
+    {
+      "epoch": 4.919629810034096,
+      "grad_norm": 1.0156532526016235,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 5050
+    },
+    {
+      "epoch": 4.929371651242085,
+      "grad_norm": 0.8975168466567993,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 5060
+    },
+    {
+      "epoch": 4.939113492450073,
+      "grad_norm": 0.9787213802337646,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 5070
+    },
+    {
+      "epoch": 4.948855333658061,
+      "grad_norm": 1.0801568031311035,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 5080
+    },
+    {
+      "epoch": 4.95859717486605,
+      "grad_norm": 1.0655089616775513,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 5090
+    },
+    {
+      "epoch": 4.968339016074038,
+      "grad_norm": 0.8941320180892944,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 5100
+    },
+    {
+      "epoch": 4.978080857282026,
+      "grad_norm": 1.050621747970581,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 5110
+    },
+    {
+      "epoch": 4.987822698490015,
+      "grad_norm": 0.9724781513214111,
+      "learning_rate": 0.0002,
+      "loss": 1.3791,
+      "step": 5120
+    },
+    {
+      "epoch": 4.997564539698003,
+      "grad_norm": 0.9850538969039917,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 5130
+    },
+    {
+      "epoch": 4.9995129079396,
+      "eval_loss": 2.0824170112609863,
+      "eval_runtime": 55.592,
+      "eval_samples_per_second": 9.12,
+      "eval_steps_per_second": 1.151,
+      "step": 5132
+    },
+    {
+      "epoch": 5.007306380905991,
+      "grad_norm": 1.0096189975738525,
+      "learning_rate": 0.0002,
+      "loss": 1.037,
+      "step": 5140
+    },
+    {
+      "epoch": 5.01704822211398,
+      "grad_norm": 1.2403408288955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 5150
+    },
+    {
+      "epoch": 5.026790063321968,
+      "grad_norm": 1.1243221759796143,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 5160
+    },
+    {
+      "epoch": 5.036531904529956,
+      "grad_norm": 1.4745502471923828,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 5170
+    },
+    {
+      "epoch": 5.0462737457379445,
+      "grad_norm": 1.1913198232650757,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 5180
+    },
+    {
+      "epoch": 5.056015586945933,
+      "grad_norm": 1.2732855081558228,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 5190
+    },
+    {
+      "epoch": 5.065757428153921,
+      "grad_norm": 1.1737396717071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 5200
+    },
+    {
+      "epoch": 5.075499269361909,
+      "grad_norm": 1.4162768125534058,
+      "learning_rate": 0.0002,
+      "loss": 0.9754,
+      "step": 5210
+    },
+    {
+      "epoch": 5.085241110569898,
+      "grad_norm": 1.528274655342102,
+      "learning_rate": 0.0002,
+      "loss": 1.0333,
+      "step": 5220
+    },
+    {
+      "epoch": 5.094982951777886,
+      "grad_norm": 1.3966618776321411,
+      "learning_rate": 0.0002,
+      "loss": 1.0227,
+      "step": 5230
+    },
+    {
+      "epoch": 5.104724792985874,
+      "grad_norm": 1.3427953720092773,
+      "learning_rate": 0.0002,
+      "loss": 0.987,
+      "step": 5240
+    },
+    {
+      "epoch": 5.114466634193863,
+      "grad_norm": 1.6533905267715454,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 5250
+    },
+    {
+      "epoch": 5.124208475401851,
+      "grad_norm": 1.4114865064620972,
+      "learning_rate": 0.0002,
+      "loss": 1.0452,
+      "step": 5260
+    },
+    {
+      "epoch": 5.133950316609839,
+      "grad_norm": 1.5460708141326904,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 5270
+    },
+    {
+      "epoch": 5.143692157817828,
+      "grad_norm": 1.3491919040679932,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5280
+    },
+    {
+      "epoch": 5.153433999025816,
+      "grad_norm": 1.2208969593048096,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 5290
+    },
+    {
+      "epoch": 5.163175840233804,
+      "grad_norm": 1.1141403913497925,
+      "learning_rate": 0.0002,
+      "loss": 1.0362,
+      "step": 5300
+    },
+    {
+      "epoch": 5.172917681441793,
+      "grad_norm": 1.2938064336776733,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 5310
+    },
+    {
+      "epoch": 5.1826595226497805,
+      "grad_norm": 1.2704918384552002,
+      "learning_rate": 0.0002,
+      "loss": 1.0438,
+      "step": 5320
+    },
+    {
+      "epoch": 5.192401363857769,
+      "grad_norm": 1.3928544521331787,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 5330
+    },
+    {
+      "epoch": 5.202143205065758,
+      "grad_norm": 1.1993824243545532,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 5340
+    },
+    {
+      "epoch": 5.211885046273745,
+      "grad_norm": 1.5913670063018799,
+      "learning_rate": 0.0002,
+      "loss": 1.0195,
+      "step": 5350
+    },
+    {
+      "epoch": 5.221626887481734,
+      "grad_norm": 1.1577855348587036,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 5360
+    },
+    {
+      "epoch": 5.231368728689723,
+      "grad_norm": 1.4535993337631226,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 5370
+    },
+    {
+      "epoch": 5.24111056989771,
+      "grad_norm": 1.5068976879119873,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 5380
+    },
+    {
+      "epoch": 5.250852411105699,
+      "grad_norm": 1.2365459203720093,
+      "learning_rate": 0.0002,
+      "loss": 1.0068,
+      "step": 5390
+    },
+    {
+      "epoch": 5.2605942523136875,
+      "grad_norm": 1.3197922706604004,
+      "learning_rate": 0.0002,
+      "loss": 1.0145,
+      "step": 5400
+    },
+    {
+      "epoch": 5.270336093521675,
+      "grad_norm": 1.2395117282867432,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 5410
+    },
+    {
+      "epoch": 5.280077934729664,
+      "grad_norm": 1.1841236352920532,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 5420
+    },
+    {
+      "epoch": 5.289819775937652,
+      "grad_norm": 1.218003749847412,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 5430
+    },
+    {
+      "epoch": 5.29956161714564,
+      "grad_norm": 1.2210947275161743,
+      "learning_rate": 0.0002,
+      "loss": 1.0093,
+      "step": 5440
+    },
+    {
+      "epoch": 5.309303458353629,
+      "grad_norm": 1.266006588935852,
+      "learning_rate": 0.0002,
+      "loss": 0.9619,
+      "step": 5450
+    },
+    {
+      "epoch": 5.319045299561617,
+      "grad_norm": 1.2598075866699219,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 5460
+    },
+    {
+      "epoch": 5.328787140769606,
+      "grad_norm": 1.2410019636154175,
+      "learning_rate": 0.0002,
+      "loss": 1.0929,
+      "step": 5470
+    },
+    {
+      "epoch": 5.338528981977594,
+      "grad_norm": 1.249698519706726,
+      "learning_rate": 0.0002,
+      "loss": 1.058,
+      "step": 5480
+    },
+    {
+      "epoch": 5.348270823185582,
+      "grad_norm": 1.2398173809051514,
+      "learning_rate": 0.0002,
+      "loss": 1.0457,
+      "step": 5490
+    },
+    {
+      "epoch": 5.35801266439357,
+      "grad_norm": 1.2416654825210571,
+      "learning_rate": 0.0002,
+      "loss": 1.0139,
+      "step": 5500
+    },
+    {
+      "epoch": 5.3677545056015585,
+      "grad_norm": 1.398706316947937,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 5510
+    },
+    {
+      "epoch": 5.377496346809547,
+      "grad_norm": 1.3049418926239014,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 5520
+    },
+    {
+      "epoch": 5.387238188017536,
+      "grad_norm": 1.2528893947601318,
+      "learning_rate": 0.0002,
+      "loss": 1.0912,
+      "step": 5530
+    },
+    {
+      "epoch": 5.3969800292255234,
+      "grad_norm": 1.2963255643844604,
+      "learning_rate": 0.0002,
+      "loss": 1.0619,
+      "step": 5540
+    },
+    {
+      "epoch": 5.406721870433512,
+      "grad_norm": 1.494231104850769,
+      "learning_rate": 0.0002,
+      "loss": 1.0194,
+      "step": 5550
+    },
+    {
+      "epoch": 5.416463711641501,
+      "grad_norm": 1.2760992050170898,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 5560
+    },
+    {
+      "epoch": 5.426205552849488,
+      "grad_norm": 1.195292592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.1088,
+      "step": 5570
+    },
+    {
+      "epoch": 5.435947394057477,
+      "grad_norm": 1.6408965587615967,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 5580
+    },
+    {
+      "epoch": 5.4456892352654656,
+      "grad_norm": 1.3092058897018433,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 5590
+    },
+    {
+      "epoch": 5.455431076473453,
+      "grad_norm": 1.2960586547851562,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 5600
+    },
+    {
+      "epoch": 5.465172917681442,
+      "grad_norm": 1.3560487031936646,
+      "learning_rate": 0.0002,
+      "loss": 1.0257,
+      "step": 5610
+    },
+    {
+      "epoch": 5.4749147588894305,
+      "grad_norm": 1.1896311044692993,
+      "learning_rate": 0.0002,
+      "loss": 1.0314,
+      "step": 5620
+    },
+    {
+      "epoch": 5.484656600097418,
+      "grad_norm": 1.3145595788955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0435,
+      "step": 5630
+    },
+    {
+      "epoch": 5.494398441305407,
+      "grad_norm": 1.2207404375076294,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 5640
+    },
+    {
+      "epoch": 5.504140282513395,
+      "grad_norm": 1.266015887260437,
+      "learning_rate": 0.0002,
+      "loss": 1.0823,
+      "step": 5650
+    },
+    {
+      "epoch": 5.513882123721383,
+      "grad_norm": 1.2478289604187012,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 5660
+    },
+    {
+      "epoch": 5.523623964929372,
+      "grad_norm": 1.4851372241973877,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 5670
+    },
+    {
+      "epoch": 5.53336580613736,
+      "grad_norm": 1.4478679895401,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5680
+    },
+    {
+      "epoch": 5.543107647345348,
+      "grad_norm": 1.1079537868499756,
+      "learning_rate": 0.0002,
+      "loss": 1.043,
+      "step": 5690
+    },
+    {
+      "epoch": 5.552849488553337,
+      "grad_norm": 1.4201879501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 5700
+    },
+    {
+      "epoch": 5.562591329761325,
+      "grad_norm": 1.2092000246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.0697,
+      "step": 5710
+    },
+    {
+      "epoch": 5.572333170969313,
+      "grad_norm": 1.4515851736068726,
+      "learning_rate": 0.0002,
+      "loss": 0.9868,
+      "step": 5720
+    },
+    {
+      "epoch": 5.5820750121773015,
+      "grad_norm": 1.3260412216186523,
+      "learning_rate": 0.0002,
+      "loss": 1.1547,
+      "step": 5730
+    },
+    {
+      "epoch": 5.59181685338529,
+      "grad_norm": 1.248191475868225,
+      "learning_rate": 0.0002,
+      "loss": 1.1388,
+      "step": 5740
+    },
+    {
+      "epoch": 5.601558694593278,
+      "grad_norm": 1.2037307024002075,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 5750
+    },
+    {
+      "epoch": 5.611300535801266,
+      "grad_norm": 1.341237187385559,
+      "learning_rate": 0.0002,
+      "loss": 1.1425,
+      "step": 5760
+    },
+    {
+      "epoch": 5.621042377009255,
+      "grad_norm": 1.130115270614624,
+      "learning_rate": 0.0002,
+      "loss": 1.0942,
+      "step": 5770
+    },
+    {
+      "epoch": 5.630784218217243,
+      "grad_norm": 1.3834772109985352,
+      "learning_rate": 0.0002,
+      "loss": 1.1029,
+      "step": 5780
+    },
+    {
+      "epoch": 5.640526059425231,
+      "grad_norm": 1.2586270570755005,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 5790
+    },
+    {
+      "epoch": 5.65026790063322,
+      "grad_norm": 1.3233023881912231,
+      "learning_rate": 0.0002,
+      "loss": 1.0186,
+      "step": 5800
+    },
+    {
+      "epoch": 5.660009741841208,
+      "grad_norm": 1.2711341381072998,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 5810
+    },
+    {
+      "epoch": 5.669751583049196,
+      "grad_norm": 1.3867720365524292,
+      "learning_rate": 0.0002,
+      "loss": 1.0897,
+      "step": 5820
+    },
+    {
+      "epoch": 5.679493424257185,
+      "grad_norm": 1.4783269166946411,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 5830
+    },
+    {
+      "epoch": 5.6892352654651726,
+      "grad_norm": 1.2744768857955933,
+      "learning_rate": 0.0002,
+      "loss": 1.0632,
+      "step": 5840
+    },
+    {
+      "epoch": 5.698977106673161,
+      "grad_norm": 1.3405882120132446,
+      "learning_rate": 0.0002,
+      "loss": 1.1484,
+      "step": 5850
+    },
+    {
+      "epoch": 5.70871894788115,
+      "grad_norm": 1.204300880432129,
+      "learning_rate": 0.0002,
+      "loss": 1.0975,
+      "step": 5860
+    },
+    {
+      "epoch": 5.7184607890891375,
+      "grad_norm": 1.2954572439193726,
+      "learning_rate": 0.0002,
+      "loss": 1.0494,
+      "step": 5870
+    },
+    {
+      "epoch": 5.728202630297126,
+      "grad_norm": 1.5478382110595703,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 5880
+    },
+    {
+      "epoch": 5.737944471505115,
+      "grad_norm": 1.2095842361450195,
+      "learning_rate": 0.0002,
+      "loss": 1.0582,
+      "step": 5890
+    },
+    {
+      "epoch": 5.747686312713103,
+      "grad_norm": 1.0691519975662231,
+      "learning_rate": 0.0002,
+      "loss": 1.1,
+      "step": 5900
+    },
+    {
+      "epoch": 5.757428153921091,
+      "grad_norm": 1.1920677423477173,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 5910
+    },
+    {
+      "epoch": 5.76716999512908,
+      "grad_norm": 1.2051277160644531,
+      "learning_rate": 0.0002,
+      "loss": 1.1746,
+      "step": 5920
+    },
+    {
+      "epoch": 5.776911836337067,
+      "grad_norm": 1.197490930557251,
+      "learning_rate": 0.0002,
+      "loss": 1.1221,
+      "step": 5930
+    },
+    {
+      "epoch": 5.786653677545056,
+      "grad_norm": 1.2003998756408691,
+      "learning_rate": 0.0002,
+      "loss": 1.07,
+      "step": 5940
+    },
+    {
+      "epoch": 5.7963955187530445,
+      "grad_norm": 1.2323646545410156,
+      "learning_rate": 0.0002,
+      "loss": 1.0938,
+      "step": 5950
+    },
+    {
+      "epoch": 5.806137359961033,
+      "grad_norm": 1.2593932151794434,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 5960
+    },
+    {
+      "epoch": 5.815879201169021,
+      "grad_norm": 1.1835976839065552,
+      "learning_rate": 0.0002,
+      "loss": 1.0829,
+      "step": 5970
+    },
+    {
+      "epoch": 5.825621042377009,
+      "grad_norm": 1.4770104885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 5980
+    },
+    {
+      "epoch": 5.835362883584997,
+      "grad_norm": 1.1025809049606323,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 5990
+    },
+    {
+      "epoch": 5.845104724792986,
+      "grad_norm": 1.364588975906372,
+      "learning_rate": 0.0002,
+      "loss": 1.1323,
+      "step": 6000
+    },
+    {
+      "epoch": 5.854846566000974,
+      "grad_norm": 1.2340112924575806,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 6010
+    },
+    {
+      "epoch": 5.864588407208963,
+      "grad_norm": 1.4925711154937744,
+      "learning_rate": 0.0002,
+      "loss": 1.1123,
+      "step": 6020
+    },
+    {
+      "epoch": 5.874330248416951,
+      "grad_norm": 1.3516744375228882,
+      "learning_rate": 0.0002,
+      "loss": 1.12,
+      "step": 6030
+    },
+    {
+      "epoch": 5.884072089624939,
+      "grad_norm": 1.2058138847351074,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 6040
+    },
+    {
+      "epoch": 5.893813930832927,
+      "grad_norm": 1.13870108127594,
+      "learning_rate": 0.0002,
+      "loss": 1.1074,
+      "step": 6050
+    },
+    {
+      "epoch": 5.9035557720409155,
+      "grad_norm": 1.1587319374084473,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 6060
+    },
+    {
+      "epoch": 5.913297613248904,
+      "grad_norm": 1.164481520652771,
+      "learning_rate": 0.0002,
+      "loss": 1.1376,
+      "step": 6070
+    },
+    {
+      "epoch": 5.923039454456893,
+      "grad_norm": 1.2115206718444824,
+      "learning_rate": 0.0002,
+      "loss": 1.1262,
+      "step": 6080
+    },
+    {
+      "epoch": 5.93278129566488,
+      "grad_norm": 1.3201590776443481,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 6090
+    },
+    {
+      "epoch": 5.942523136872869,
+      "grad_norm": 1.287380576133728,
+      "learning_rate": 0.0002,
+      "loss": 1.1288,
+      "step": 6100
+    },
+    {
+      "epoch": 5.952264978080858,
+      "grad_norm": 1.1820166110992432,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 6110
+    },
+    {
+      "epoch": 5.962006819288845,
+      "grad_norm": 1.2550667524337769,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 6120
+    },
+    {
+      "epoch": 5.971748660496834,
+      "grad_norm": 1.3547813892364502,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 6130
+    },
+    {
+      "epoch": 5.9814905017048225,
+      "grad_norm": 1.260842204093933,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 6140
+    },
+    {
+      "epoch": 5.99123234291281,
+      "grad_norm": 1.1643036603927612,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 6150
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.2628161907196045,
+      "eval_runtime": 57.2379,
+      "eval_samples_per_second": 8.858,
+      "eval_steps_per_second": 1.118,
+      "step": 6159
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 3.1634930932383744e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-6159/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a476254c5476ecf662100e81977dbff8e31af0b8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ad21b406fa7ff475f2aacc870dc56a79db5aa11115489ef6c48349e78b13267
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d0b3a7055a8064cadb3433d54e50ff81e4c28b2e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adb60d96f48eec71564d7294f0ee00a6eee0d4c8fc2832b91a65336aed1c90bc
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dd39b7bf319198de21b9e2a498f3818eaf87d824
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c177f26bb6249de2a1cc5290b39f6a754b9ac519df5f3be6f7e8066aa6640f0
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0512da751f5b1afd057bc1deacf2719d3925e537
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e4edaec93b10616e7d2a79bcf0bd5b1c26686d61184d25e7d0a4b9ea9617362
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..76679bf68796a18e3b82a06687d1b36221aae8b2
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/trainer_state.json
@@ -0,0 +1,5115 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 6.9995129079396,
+  "eval_steps": 10,
+  "global_step": 7185,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    },
+    {
+      "epoch": 3.0004870920603994,
+      "grad_norm": 0.49992576241493225,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 3.0102289332683876,
+      "grad_norm": 0.8242783546447754,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3090
+    },
+    {
+      "epoch": 3.019970774476376,
+      "grad_norm": 0.6330569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3100
+    },
+    {
+      "epoch": 3.0297126156843643,
+      "grad_norm": 0.566097617149353,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 3110
+    },
+    {
+      "epoch": 3.0394544568923525,
+      "grad_norm": 0.6337586045265198,
+      "learning_rate": 0.0002,
+      "loss": 1.4365,
+      "step": 3120
+    },
+    {
+      "epoch": 3.049196298100341,
+      "grad_norm": 0.7339403033256531,
+      "learning_rate": 0.0002,
+      "loss": 1.3916,
+      "step": 3130
+    },
+    {
+      "epoch": 3.0589381393083293,
+      "grad_norm": 0.7187346816062927,
+      "learning_rate": 0.0002,
+      "loss": 1.4617,
+      "step": 3140
+    },
+    {
+      "epoch": 3.0686799805163174,
+      "grad_norm": 0.7116255760192871,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 3150
+    },
+    {
+      "epoch": 3.078421821724306,
+      "grad_norm": 0.6493807435035706,
+      "learning_rate": 0.0002,
+      "loss": 1.4452,
+      "step": 3160
+    },
+    {
+      "epoch": 3.088163662932294,
+      "grad_norm": 0.6777266263961792,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 3170
+    },
+    {
+      "epoch": 3.0979055041402823,
+      "grad_norm": 0.6342006325721741,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 3180
+    },
+    {
+      "epoch": 3.107647345348271,
+      "grad_norm": 0.6608964204788208,
+      "learning_rate": 0.0002,
+      "loss": 1.4748,
+      "step": 3190
+    },
+    {
+      "epoch": 3.117389186556259,
+      "grad_norm": 0.7230247259140015,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 3200
+    },
+    {
+      "epoch": 3.1271310277642472,
+      "grad_norm": 0.650368332862854,
+      "learning_rate": 0.0002,
+      "loss": 1.4049,
+      "step": 3210
+    },
+    {
+      "epoch": 3.136872868972236,
+      "grad_norm": 0.7319342494010925,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 3220
+    },
+    {
+      "epoch": 3.146614710180224,
+      "grad_norm": 0.7159963846206665,
+      "learning_rate": 0.0002,
+      "loss": 1.3872,
+      "step": 3230
+    },
+    {
+      "epoch": 3.156356551388212,
+      "grad_norm": 0.8905230164527893,
+      "learning_rate": 0.0002,
+      "loss": 1.5076,
+      "step": 3240
+    },
+    {
+      "epoch": 3.1660983925962007,
+      "grad_norm": 0.6920804381370544,
+      "learning_rate": 0.0002,
+      "loss": 1.3161,
+      "step": 3250
+    },
+    {
+      "epoch": 3.175840233804189,
+      "grad_norm": 0.6782063841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.3786,
+      "step": 3260
+    },
+    {
+      "epoch": 3.1855820750121775,
+      "grad_norm": 0.735325276851654,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 3270
+    },
+    {
+      "epoch": 3.1953239162201656,
+      "grad_norm": 0.6657978296279907,
+      "learning_rate": 0.0002,
+      "loss": 1.4027,
+      "step": 3280
+    },
+    {
+      "epoch": 3.205065757428154,
+      "grad_norm": 0.771315336227417,
+      "learning_rate": 0.0002,
+      "loss": 1.3456,
+      "step": 3290
+    },
+    {
+      "epoch": 3.2148075986361424,
+      "grad_norm": 0.6492983102798462,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 3300
+    },
+    {
+      "epoch": 3.2245494398441306,
+      "grad_norm": 0.7513770461082458,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3310
+    },
+    {
+      "epoch": 3.2342912810521187,
+      "grad_norm": 0.7091423869132996,
+      "learning_rate": 0.0002,
+      "loss": 1.4032,
+      "step": 3320
+    },
+    {
+      "epoch": 3.2440331222601073,
+      "grad_norm": 0.6663975119590759,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 3330
+    },
+    {
+      "epoch": 3.2537749634680955,
+      "grad_norm": 0.6813122034072876,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 3340
+    },
+    {
+      "epoch": 3.2635168046760836,
+      "grad_norm": 0.6602569818496704,
+      "learning_rate": 0.0002,
+      "loss": 1.3681,
+      "step": 3350
+    },
+    {
+      "epoch": 3.2732586458840722,
+      "grad_norm": 0.718270480632782,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 3360
+    },
+    {
+      "epoch": 3.2830004870920604,
+      "grad_norm": 0.6884173154830933,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 3370
+    },
+    {
+      "epoch": 3.2927423283000485,
+      "grad_norm": 0.7039775848388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3380
+    },
+    {
+      "epoch": 3.302484169508037,
+      "grad_norm": 0.7444299459457397,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3390
+    },
+    {
+      "epoch": 3.3122260107160253,
+      "grad_norm": 0.7187064290046692,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 3400
+    },
+    {
+      "epoch": 3.3219678519240134,
+      "grad_norm": 0.599396288394928,
+      "learning_rate": 0.0002,
+      "loss": 1.3684,
+      "step": 3410
+    },
+    {
+      "epoch": 3.331709693132002,
+      "grad_norm": 0.7670390009880066,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3420
+    },
+    {
+      "epoch": 3.34145153433999,
+      "grad_norm": 0.6654478311538696,
+      "learning_rate": 0.0002,
+      "loss": 1.4411,
+      "step": 3430
+    },
+    {
+      "epoch": 3.351193375547979,
+      "grad_norm": 0.6644385457038879,
+      "learning_rate": 0.0002,
+      "loss": 1.4257,
+      "step": 3440
+    },
+    {
+      "epoch": 3.360935216755967,
+      "grad_norm": 0.6974098086357117,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3450
+    },
+    {
+      "epoch": 3.370677057963955,
+      "grad_norm": 0.7350399494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 3460
+    },
+    {
+      "epoch": 3.3804188991719437,
+      "grad_norm": 0.714721143245697,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 3470
+    },
+    {
+      "epoch": 3.390160740379932,
+      "grad_norm": 0.7006027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 3480
+    },
+    {
+      "epoch": 3.39990258158792,
+      "grad_norm": 0.6767925024032593,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 3490
+    },
+    {
+      "epoch": 3.4096444227959086,
+      "grad_norm": 0.6721355319023132,
+      "learning_rate": 0.0002,
+      "loss": 1.4116,
+      "step": 3500
+    },
+    {
+      "epoch": 3.419386264003897,
+      "grad_norm": 0.6845725178718567,
+      "learning_rate": 0.0002,
+      "loss": 1.443,
+      "step": 3510
+    },
+    {
+      "epoch": 3.429128105211885,
+      "grad_norm": 0.6882196664810181,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 3520
+    },
+    {
+      "epoch": 3.4388699464198735,
+      "grad_norm": 0.7663240432739258,
+      "learning_rate": 0.0002,
+      "loss": 1.4962,
+      "step": 3530
+    },
+    {
+      "epoch": 3.4486117876278617,
+      "grad_norm": 0.6304219365119934,
+      "learning_rate": 0.0002,
+      "loss": 1.4644,
+      "step": 3540
+    },
+    {
+      "epoch": 3.45835362883585,
+      "grad_norm": 0.668678879737854,
+      "learning_rate": 0.0002,
+      "loss": 1.4918,
+      "step": 3550
+    },
+    {
+      "epoch": 3.4680954700438384,
+      "grad_norm": 0.7526912093162537,
+      "learning_rate": 0.0002,
+      "loss": 1.4874,
+      "step": 3560
+    },
+    {
+      "epoch": 3.4778373112518266,
+      "grad_norm": 1.089495301246643,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 3570
+    },
+    {
+      "epoch": 3.4875791524598148,
+      "grad_norm": 0.7282902002334595,
+      "learning_rate": 0.0002,
+      "loss": 1.3871,
+      "step": 3580
+    },
+    {
+      "epoch": 3.4973209936678034,
+      "grad_norm": 0.6540156602859497,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3590
+    },
+    {
+      "epoch": 3.5070628348757915,
+      "grad_norm": 0.6449568867683411,
+      "learning_rate": 0.0002,
+      "loss": 1.4367,
+      "step": 3600
+    },
+    {
+      "epoch": 3.5168046760837797,
+      "grad_norm": 0.7262216210365295,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 3610
+    },
+    {
+      "epoch": 3.5265465172917683,
+      "grad_norm": 0.6048615574836731,
+      "learning_rate": 0.0002,
+      "loss": 1.4374,
+      "step": 3620
+    },
+    {
+      "epoch": 3.5362883584997564,
+      "grad_norm": 0.6780537366867065,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 3630
+    },
+    {
+      "epoch": 3.5460301997077446,
+      "grad_norm": 0.6851925253868103,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 3640
+    },
+    {
+      "epoch": 3.555772040915733,
+      "grad_norm": 0.6530634164810181,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 3650
+    },
+    {
+      "epoch": 3.5655138821237213,
+      "grad_norm": 0.7193992733955383,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 3660
+    },
+    {
+      "epoch": 3.5752557233317095,
+      "grad_norm": 0.767496645450592,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 3670
+    },
+    {
+      "epoch": 3.584997564539698,
+      "grad_norm": 0.6912919282913208,
+      "learning_rate": 0.0002,
+      "loss": 1.4824,
+      "step": 3680
+    },
+    {
+      "epoch": 3.5947394057476862,
+      "grad_norm": 0.7383436560630798,
+      "learning_rate": 0.0002,
+      "loss": 1.4497,
+      "step": 3690
+    },
+    {
+      "epoch": 3.6044812469556744,
+      "grad_norm": 0.6746662855148315,
+      "learning_rate": 0.0002,
+      "loss": 1.4822,
+      "step": 3700
+    },
+    {
+      "epoch": 3.614223088163663,
+      "grad_norm": 0.6885138750076294,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 3710
+    },
+    {
+      "epoch": 3.623964929371651,
+      "grad_norm": 0.6694392561912537,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 3720
+    },
+    {
+      "epoch": 3.6337067705796393,
+      "grad_norm": 0.812358021736145,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 3730
+    },
+    {
+      "epoch": 3.643448611787628,
+      "grad_norm": 0.7267130017280579,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 3740
+    },
+    {
+      "epoch": 3.653190452995616,
+      "grad_norm": 0.6958749294281006,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 3750
+    },
+    {
+      "epoch": 3.6629322942036042,
+      "grad_norm": 0.6805673241615295,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3760
+    },
+    {
+      "epoch": 3.672674135411593,
+      "grad_norm": 0.7184410095214844,
+      "learning_rate": 0.0002,
+      "loss": 1.4338,
+      "step": 3770
+    },
+    {
+      "epoch": 3.682415976619581,
+      "grad_norm": 0.7716330289840698,
+      "learning_rate": 0.0002,
+      "loss": 1.3935,
+      "step": 3780
+    },
+    {
+      "epoch": 3.6921578178275696,
+      "grad_norm": 0.6675831079483032,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 3790
+    },
+    {
+      "epoch": 3.7018996590355577,
+      "grad_norm": 0.6480095386505127,
+      "learning_rate": 0.0002,
+      "loss": 1.401,
+      "step": 3800
+    },
+    {
+      "epoch": 3.711641500243546,
+      "grad_norm": 0.6559418439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 3810
+    },
+    {
+      "epoch": 3.7213833414515345,
+      "grad_norm": 0.6596545577049255,
+      "learning_rate": 0.0002,
+      "loss": 1.4341,
+      "step": 3820
+    },
+    {
+      "epoch": 3.7311251826595226,
+      "grad_norm": 0.7172950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3830
+    },
+    {
+      "epoch": 3.740867023867511,
+      "grad_norm": 0.796148419380188,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 3840
+    },
+    {
+      "epoch": 3.7506088650754994,
+      "grad_norm": 0.6600322723388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4992,
+      "step": 3850
+    },
+    {
+      "epoch": 3.7603507062834876,
+      "grad_norm": 0.6776387691497803,
+      "learning_rate": 0.0002,
+      "loss": 1.4201,
+      "step": 3860
+    },
+    {
+      "epoch": 3.770092547491476,
+      "grad_norm": 0.7768304347991943,
+      "learning_rate": 0.0002,
+      "loss": 1.3893,
+      "step": 3870
+    },
+    {
+      "epoch": 3.7798343886994643,
+      "grad_norm": 1.0579794645309448,
+      "learning_rate": 0.0002,
+      "loss": 1.4886,
+      "step": 3880
+    },
+    {
+      "epoch": 3.7895762299074525,
+      "grad_norm": 0.6757252812385559,
+      "learning_rate": 0.0002,
+      "loss": 1.4556,
+      "step": 3890
+    },
+    {
+      "epoch": 3.799318071115441,
+      "grad_norm": 0.6706996560096741,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 3900
+    },
+    {
+      "epoch": 3.809059912323429,
+      "grad_norm": 0.7026948928833008,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 3910
+    },
+    {
+      "epoch": 3.8188017535314174,
+      "grad_norm": 0.6437768340110779,
+      "learning_rate": 0.0002,
+      "loss": 1.5487,
+      "step": 3920
+    },
+    {
+      "epoch": 3.828543594739406,
+      "grad_norm": 0.7015706300735474,
+      "learning_rate": 0.0002,
+      "loss": 1.4678,
+      "step": 3930
+    },
+    {
+      "epoch": 3.838285435947394,
+      "grad_norm": 0.7049482464790344,
+      "learning_rate": 0.0002,
+      "loss": 1.4891,
+      "step": 3940
+    },
+    {
+      "epoch": 3.8480272771553823,
+      "grad_norm": 0.6533724665641785,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 3950
+    },
+    {
+      "epoch": 3.857769118363371,
+      "grad_norm": 0.7312499284744263,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 3960
+    },
+    {
+      "epoch": 3.867510959571359,
+      "grad_norm": 0.6858801245689392,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3970
+    },
+    {
+      "epoch": 3.877252800779347,
+      "grad_norm": 0.770423173904419,
+      "learning_rate": 0.0002,
+      "loss": 1.4423,
+      "step": 3980
+    },
+    {
+      "epoch": 3.886994641987336,
+      "grad_norm": 0.6987539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 3990
+    },
+    {
+      "epoch": 3.896736483195324,
+      "grad_norm": 0.7072722315788269,
+      "learning_rate": 0.0002,
+      "loss": 1.4791,
+      "step": 4000
+    },
+    {
+      "epoch": 3.906478324403312,
+      "grad_norm": 0.6492931842803955,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4010
+    },
+    {
+      "epoch": 3.9162201656113007,
+      "grad_norm": 0.7716232538223267,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 4020
+    },
+    {
+      "epoch": 3.925962006819289,
+      "grad_norm": 0.722949743270874,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 4030
+    },
+    {
+      "epoch": 3.935703848027277,
+      "grad_norm": 0.7434365749359131,
+      "learning_rate": 0.0002,
+      "loss": 1.3914,
+      "step": 4040
+    },
+    {
+      "epoch": 3.9454456892352656,
+      "grad_norm": 0.6691509485244751,
+      "learning_rate": 0.0002,
+      "loss": 1.4763,
+      "step": 4050
+    },
+    {
+      "epoch": 3.9551875304432538,
+      "grad_norm": 0.6850284337997437,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 4060
+    },
+    {
+      "epoch": 3.964929371651242,
+      "grad_norm": 0.6954452991485596,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4070
+    },
+    {
+      "epoch": 3.9746712128592305,
+      "grad_norm": 0.9316364526748657,
+      "learning_rate": 0.0002,
+      "loss": 1.417,
+      "step": 4080
+    },
+    {
+      "epoch": 3.9844130540672187,
+      "grad_norm": 0.6908289194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 4090
+    },
+    {
+      "epoch": 3.994154895275207,
+      "grad_norm": 0.666782021522522,
+      "learning_rate": 0.0002,
+      "loss": 1.4404,
+      "step": 4100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9233275651931763,
+      "eval_runtime": 55.9536,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 1.144,
+      "step": 4106
+    },
+    {
+      "epoch": 4.003896736483195,
+      "grad_norm": 0.7726166248321533,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 4110
+    },
+    {
+      "epoch": 4.013638577691184,
+      "grad_norm": 1.1338967084884644,
+      "learning_rate": 0.0002,
+      "loss": 1.1415,
+      "step": 4120
+    },
+    {
+      "epoch": 4.023380418899172,
+      "grad_norm": 0.9530029296875,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 4130
+    },
+    {
+      "epoch": 4.03312226010716,
+      "grad_norm": 1.1058554649353027,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 4140
+    },
+    {
+      "epoch": 4.042864101315149,
+      "grad_norm": 0.8765049576759338,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 4150
+    },
+    {
+      "epoch": 4.052605942523137,
+      "grad_norm": 1.1774667501449585,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4160
+    },
+    {
+      "epoch": 4.062347783731125,
+      "grad_norm": 0.9301433563232422,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 4170
+    },
+    {
+      "epoch": 4.072089624939114,
+      "grad_norm": 1.0196778774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.1807,
+      "step": 4180
+    },
+    {
+      "epoch": 4.081831466147102,
+      "grad_norm": 1.1380577087402344,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 4190
+    },
+    {
+      "epoch": 4.09157330735509,
+      "grad_norm": 0.9121319651603699,
+      "learning_rate": 0.0002,
+      "loss": 1.2521,
+      "step": 4200
+    },
+    {
+      "epoch": 4.101315148563079,
+      "grad_norm": 0.9495378732681274,
+      "learning_rate": 0.0002,
+      "loss": 1.1747,
+      "step": 4210
+    },
+    {
+      "epoch": 4.1110569897710665,
+      "grad_norm": 0.8058680295944214,
+      "learning_rate": 0.0002,
+      "loss": 1.1829,
+      "step": 4220
+    },
+    {
+      "epoch": 4.120798830979055,
+      "grad_norm": 1.000887393951416,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 4230
+    },
+    {
+      "epoch": 4.130540672187044,
+      "grad_norm": 0.9529102444648743,
+      "learning_rate": 0.0002,
+      "loss": 1.1947,
+      "step": 4240
+    },
+    {
+      "epoch": 4.140282513395031,
+      "grad_norm": 1.0257115364074707,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 4250
+    },
+    {
+      "epoch": 4.15002435460302,
+      "grad_norm": 0.9590303897857666,
+      "learning_rate": 0.0002,
+      "loss": 1.2293,
+      "step": 4260
+    },
+    {
+      "epoch": 4.159766195811009,
+      "grad_norm": 1.065291166305542,
+      "learning_rate": 0.0002,
+      "loss": 1.1918,
+      "step": 4270
+    },
+    {
+      "epoch": 4.169508037018996,
+      "grad_norm": 0.8819697499275208,
+      "learning_rate": 0.0002,
+      "loss": 1.2323,
+      "step": 4280
+    },
+    {
+      "epoch": 4.179249878226985,
+      "grad_norm": 1.0335261821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 4290
+    },
+    {
+      "epoch": 4.1889917194349735,
+      "grad_norm": 0.8872809410095215,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 4300
+    },
+    {
+      "epoch": 4.198733560642961,
+      "grad_norm": 0.9883159399032593,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 4310
+    },
+    {
+      "epoch": 4.20847540185095,
+      "grad_norm": 1.0254192352294922,
+      "learning_rate": 0.0002,
+      "loss": 1.2544,
+      "step": 4320
+    },
+    {
+      "epoch": 4.218217243058938,
+      "grad_norm": 0.9432600736618042,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 4330
+    },
+    {
+      "epoch": 4.227959084266926,
+      "grad_norm": 1.1008676290512085,
+      "learning_rate": 0.0002,
+      "loss": 1.2684,
+      "step": 4340
+    },
+    {
+      "epoch": 4.237700925474915,
+      "grad_norm": 1.0829699039459229,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 4350
+    },
+    {
+      "epoch": 4.247442766682903,
+      "grad_norm": 1.016847848892212,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4360
+    },
+    {
+      "epoch": 4.257184607890891,
+      "grad_norm": 0.8924864530563354,
+      "learning_rate": 0.0002,
+      "loss": 1.2375,
+      "step": 4370
+    },
+    {
+      "epoch": 4.26692644909888,
+      "grad_norm": 0.9300530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.1987,
+      "step": 4380
+    },
+    {
+      "epoch": 4.276668290306868,
+      "grad_norm": 0.9684814810752869,
+      "learning_rate": 0.0002,
+      "loss": 1.1696,
+      "step": 4390
+    },
+    {
+      "epoch": 4.286410131514856,
+      "grad_norm": 0.9916250705718994,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 4400
+    },
+    {
+      "epoch": 4.2961519727228445,
+      "grad_norm": 0.903680145740509,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 4410
+    },
+    {
+      "epoch": 4.305893813930833,
+      "grad_norm": 0.8713505268096924,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 4420
+    },
+    {
+      "epoch": 4.315635655138821,
+      "grad_norm": 0.9983905553817749,
+      "learning_rate": 0.0002,
+      "loss": 1.1957,
+      "step": 4430
+    },
+    {
+      "epoch": 4.3253774963468095,
+      "grad_norm": 1.1689040660858154,
+      "learning_rate": 0.0002,
+      "loss": 1.2676,
+      "step": 4440
+    },
+    {
+      "epoch": 4.335119337554798,
+      "grad_norm": 0.9316853880882263,
+      "learning_rate": 0.0002,
+      "loss": 1.2166,
+      "step": 4450
+    },
+    {
+      "epoch": 4.344861178762786,
+      "grad_norm": 0.9175887107849121,
+      "learning_rate": 0.0002,
+      "loss": 1.222,
+      "step": 4460
+    },
+    {
+      "epoch": 4.354603019970774,
+      "grad_norm": 0.9348906874656677,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4470
+    },
+    {
+      "epoch": 4.364344861178763,
+      "grad_norm": 0.9727016687393188,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 4480
+    },
+    {
+      "epoch": 4.374086702386751,
+      "grad_norm": 0.9843429923057556,
+      "learning_rate": 0.0002,
+      "loss": 1.2616,
+      "step": 4490
+    },
+    {
+      "epoch": 4.383828543594739,
+      "grad_norm": 0.9615852236747742,
+      "learning_rate": 0.0002,
+      "loss": 1.2488,
+      "step": 4500
+    },
+    {
+      "epoch": 4.393570384802728,
+      "grad_norm": 0.9688583612442017,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 4510
+    },
+    {
+      "epoch": 4.403312226010716,
+      "grad_norm": 0.9933668375015259,
+      "learning_rate": 0.0002,
+      "loss": 1.2546,
+      "step": 4520
+    },
+    {
+      "epoch": 4.413054067218704,
+      "grad_norm": 1.0626686811447144,
+      "learning_rate": 0.0002,
+      "loss": 1.2355,
+      "step": 4530
+    },
+    {
+      "epoch": 4.422795908426693,
+      "grad_norm": 0.9536267518997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 4540
+    },
+    {
+      "epoch": 4.432537749634681,
+      "grad_norm": 0.9777140021324158,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4550
+    },
+    {
+      "epoch": 4.442279590842669,
+      "grad_norm": 0.980780839920044,
+      "learning_rate": 0.0002,
+      "loss": 1.2878,
+      "step": 4560
+    },
+    {
+      "epoch": 4.452021432050658,
+      "grad_norm": 1.0147196054458618,
+      "learning_rate": 0.0002,
+      "loss": 1.2597,
+      "step": 4570
+    },
+    {
+      "epoch": 4.461763273258645,
+      "grad_norm": 0.9763361811637878,
+      "learning_rate": 0.0002,
+      "loss": 1.2148,
+      "step": 4580
+    },
+    {
+      "epoch": 4.471505114466634,
+      "grad_norm": 1.0300798416137695,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 4590
+    },
+    {
+      "epoch": 4.481246955674623,
+      "grad_norm": 0.8833121657371521,
+      "learning_rate": 0.0002,
+      "loss": 1.2665,
+      "step": 4600
+    },
+    {
+      "epoch": 4.490988796882611,
+      "grad_norm": 1.1214020252227783,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 4610
+    },
+    {
+      "epoch": 4.500730638090599,
+      "grad_norm": 0.8843787908554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2579,
+      "step": 4620
+    },
+    {
+      "epoch": 4.5104724792985875,
+      "grad_norm": 0.9942020773887634,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 4630
+    },
+    {
+      "epoch": 4.520214320506576,
+      "grad_norm": 1.0033202171325684,
+      "learning_rate": 0.0002,
+      "loss": 1.3172,
+      "step": 4640
+    },
+    {
+      "epoch": 4.529956161714564,
+      "grad_norm": 0.8767235279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2024,
+      "step": 4650
+    },
+    {
+      "epoch": 4.539698002922552,
+      "grad_norm": 1.0117276906967163,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 4660
+    },
+    {
+      "epoch": 4.549439844130541,
+      "grad_norm": 1.2787362337112427,
+      "learning_rate": 0.0002,
+      "loss": 1.2911,
+      "step": 4670
+    },
+    {
+      "epoch": 4.559181685338529,
+      "grad_norm": 0.8824878931045532,
+      "learning_rate": 0.0002,
+      "loss": 1.2603,
+      "step": 4680
+    },
+    {
+      "epoch": 4.568923526546517,
+      "grad_norm": 0.9209560751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4690
+    },
+    {
+      "epoch": 4.578665367754506,
+      "grad_norm": 1.1064010858535767,
+      "learning_rate": 0.0002,
+      "loss": 1.1916,
+      "step": 4700
+    },
+    {
+      "epoch": 4.588407208962494,
+      "grad_norm": 0.8914572596549988,
+      "learning_rate": 0.0002,
+      "loss": 1.2217,
+      "step": 4710
+    },
+    {
+      "epoch": 4.598149050170482,
+      "grad_norm": 1.0412265062332153,
+      "learning_rate": 0.0002,
+      "loss": 1.2861,
+      "step": 4720
+    },
+    {
+      "epoch": 4.607890891378471,
+      "grad_norm": 1.1950221061706543,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4730
+    },
+    {
+      "epoch": 4.617632732586459,
+      "grad_norm": 0.8938062787055969,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 4740
+    },
+    {
+      "epoch": 4.627374573794447,
+      "grad_norm": 0.9849569201469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4750
+    },
+    {
+      "epoch": 4.637116415002436,
+      "grad_norm": 1.0081515312194824,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 4760
+    },
+    {
+      "epoch": 4.6468582562104235,
+      "grad_norm": 0.8566309213638306,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 4770
+    },
+    {
+      "epoch": 4.656600097418412,
+      "grad_norm": 1.1750118732452393,
+      "learning_rate": 0.0002,
+      "loss": 1.2723,
+      "step": 4780
+    },
+    {
+      "epoch": 4.666341938626401,
+      "grad_norm": 0.925502598285675,
+      "learning_rate": 0.0002,
+      "loss": 1.2537,
+      "step": 4790
+    },
+    {
+      "epoch": 4.676083779834388,
+      "grad_norm": 1.0402472019195557,
+      "learning_rate": 0.0002,
+      "loss": 1.2146,
+      "step": 4800
+    },
+    {
+      "epoch": 4.685825621042377,
+      "grad_norm": 0.9772472977638245,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 4810
+    },
+    {
+      "epoch": 4.695567462250366,
+      "grad_norm": 0.9082779288291931,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 4820
+    },
+    {
+      "epoch": 4.705309303458353,
+      "grad_norm": 0.8026862740516663,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 4830
+    },
+    {
+      "epoch": 4.715051144666342,
+      "grad_norm": 1.1631089448928833,
+      "learning_rate": 0.0002,
+      "loss": 1.3369,
+      "step": 4840
+    },
+    {
+      "epoch": 4.7247929858743305,
+      "grad_norm": 0.9384787678718567,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 4850
+    },
+    {
+      "epoch": 4.734534827082318,
+      "grad_norm": 1.2151581048965454,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 4860
+    },
+    {
+      "epoch": 4.744276668290307,
+      "grad_norm": 0.9679436087608337,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 4870
+    },
+    {
+      "epoch": 4.754018509498295,
+      "grad_norm": 0.8352158069610596,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 4880
+    },
+    {
+      "epoch": 4.763760350706283,
+      "grad_norm": 1.0205804109573364,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 4890
+    },
+    {
+      "epoch": 4.773502191914272,
+      "grad_norm": 0.9814772605895996,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 4900
+    },
+    {
+      "epoch": 4.78324403312226,
+      "grad_norm": 1.002854347229004,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 4910
+    },
+    {
+      "epoch": 4.792985874330248,
+      "grad_norm": 1.1609505414962769,
+      "learning_rate": 0.0002,
+      "loss": 1.3143,
+      "step": 4920
+    },
+    {
+      "epoch": 4.802727715538237,
+      "grad_norm": 0.9354982376098633,
+      "learning_rate": 0.0002,
+      "loss": 1.3166,
+      "step": 4930
+    },
+    {
+      "epoch": 4.812469556746225,
+      "grad_norm": 0.9761685729026794,
+      "learning_rate": 0.0002,
+      "loss": 1.2978,
+      "step": 4940
+    },
+    {
+      "epoch": 4.822211397954213,
+      "grad_norm": 1.0604596138000488,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 4950
+    },
+    {
+      "epoch": 4.8319532391622015,
+      "grad_norm": 1.0902808904647827,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 4960
+    },
+    {
+      "epoch": 4.84169508037019,
+      "grad_norm": 1.0174955129623413,
+      "learning_rate": 0.0002,
+      "loss": 1.3073,
+      "step": 4970
+    },
+    {
+      "epoch": 4.851436921578179,
+      "grad_norm": 1.0995253324508667,
+      "learning_rate": 0.0002,
+      "loss": 1.3141,
+      "step": 4980
+    },
+    {
+      "epoch": 4.8611787627861665,
+      "grad_norm": 0.880993127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4990
+    },
+    {
+      "epoch": 4.870920603994155,
+      "grad_norm": 0.9472237825393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2547,
+      "step": 5000
+    },
+    {
+      "epoch": 4.880662445202143,
+      "grad_norm": 0.9504236578941345,
+      "learning_rate": 0.0002,
+      "loss": 1.4078,
+      "step": 5010
+    },
+    {
+      "epoch": 4.890404286410131,
+      "grad_norm": 1.1261742115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2791,
+      "step": 5020
+    },
+    {
+      "epoch": 4.90014612761812,
+      "grad_norm": 0.904674768447876,
+      "learning_rate": 0.0002,
+      "loss": 1.3707,
+      "step": 5030
+    },
+    {
+      "epoch": 4.909887968826109,
+      "grad_norm": 0.8828991055488586,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 5040
+    },
+    {
+      "epoch": 4.919629810034096,
+      "grad_norm": 1.0156532526016235,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 5050
+    },
+    {
+      "epoch": 4.929371651242085,
+      "grad_norm": 0.8975168466567993,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 5060
+    },
+    {
+      "epoch": 4.939113492450073,
+      "grad_norm": 0.9787213802337646,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 5070
+    },
+    {
+      "epoch": 4.948855333658061,
+      "grad_norm": 1.0801568031311035,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 5080
+    },
+    {
+      "epoch": 4.95859717486605,
+      "grad_norm": 1.0655089616775513,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 5090
+    },
+    {
+      "epoch": 4.968339016074038,
+      "grad_norm": 0.8941320180892944,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 5100
+    },
+    {
+      "epoch": 4.978080857282026,
+      "grad_norm": 1.050621747970581,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 5110
+    },
+    {
+      "epoch": 4.987822698490015,
+      "grad_norm": 0.9724781513214111,
+      "learning_rate": 0.0002,
+      "loss": 1.3791,
+      "step": 5120
+    },
+    {
+      "epoch": 4.997564539698003,
+      "grad_norm": 0.9850538969039917,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 5130
+    },
+    {
+      "epoch": 4.9995129079396,
+      "eval_loss": 2.0824170112609863,
+      "eval_runtime": 55.592,
+      "eval_samples_per_second": 9.12,
+      "eval_steps_per_second": 1.151,
+      "step": 5132
+    },
+    {
+      "epoch": 5.007306380905991,
+      "grad_norm": 1.0096189975738525,
+      "learning_rate": 0.0002,
+      "loss": 1.037,
+      "step": 5140
+    },
+    {
+      "epoch": 5.01704822211398,
+      "grad_norm": 1.2403408288955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 5150
+    },
+    {
+      "epoch": 5.026790063321968,
+      "grad_norm": 1.1243221759796143,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 5160
+    },
+    {
+      "epoch": 5.036531904529956,
+      "grad_norm": 1.4745502471923828,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 5170
+    },
+    {
+      "epoch": 5.0462737457379445,
+      "grad_norm": 1.1913198232650757,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 5180
+    },
+    {
+      "epoch": 5.056015586945933,
+      "grad_norm": 1.2732855081558228,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 5190
+    },
+    {
+      "epoch": 5.065757428153921,
+      "grad_norm": 1.1737396717071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 5200
+    },
+    {
+      "epoch": 5.075499269361909,
+      "grad_norm": 1.4162768125534058,
+      "learning_rate": 0.0002,
+      "loss": 0.9754,
+      "step": 5210
+    },
+    {
+      "epoch": 5.085241110569898,
+      "grad_norm": 1.528274655342102,
+      "learning_rate": 0.0002,
+      "loss": 1.0333,
+      "step": 5220
+    },
+    {
+      "epoch": 5.094982951777886,
+      "grad_norm": 1.3966618776321411,
+      "learning_rate": 0.0002,
+      "loss": 1.0227,
+      "step": 5230
+    },
+    {
+      "epoch": 5.104724792985874,
+      "grad_norm": 1.3427953720092773,
+      "learning_rate": 0.0002,
+      "loss": 0.987,
+      "step": 5240
+    },
+    {
+      "epoch": 5.114466634193863,
+      "grad_norm": 1.6533905267715454,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 5250
+    },
+    {
+      "epoch": 5.124208475401851,
+      "grad_norm": 1.4114865064620972,
+      "learning_rate": 0.0002,
+      "loss": 1.0452,
+      "step": 5260
+    },
+    {
+      "epoch": 5.133950316609839,
+      "grad_norm": 1.5460708141326904,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 5270
+    },
+    {
+      "epoch": 5.143692157817828,
+      "grad_norm": 1.3491919040679932,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5280
+    },
+    {
+      "epoch": 5.153433999025816,
+      "grad_norm": 1.2208969593048096,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 5290
+    },
+    {
+      "epoch": 5.163175840233804,
+      "grad_norm": 1.1141403913497925,
+      "learning_rate": 0.0002,
+      "loss": 1.0362,
+      "step": 5300
+    },
+    {
+      "epoch": 5.172917681441793,
+      "grad_norm": 1.2938064336776733,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 5310
+    },
+    {
+      "epoch": 5.1826595226497805,
+      "grad_norm": 1.2704918384552002,
+      "learning_rate": 0.0002,
+      "loss": 1.0438,
+      "step": 5320
+    },
+    {
+      "epoch": 5.192401363857769,
+      "grad_norm": 1.3928544521331787,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 5330
+    },
+    {
+      "epoch": 5.202143205065758,
+      "grad_norm": 1.1993824243545532,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 5340
+    },
+    {
+      "epoch": 5.211885046273745,
+      "grad_norm": 1.5913670063018799,
+      "learning_rate": 0.0002,
+      "loss": 1.0195,
+      "step": 5350
+    },
+    {
+      "epoch": 5.221626887481734,
+      "grad_norm": 1.1577855348587036,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 5360
+    },
+    {
+      "epoch": 5.231368728689723,
+      "grad_norm": 1.4535993337631226,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 5370
+    },
+    {
+      "epoch": 5.24111056989771,
+      "grad_norm": 1.5068976879119873,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 5380
+    },
+    {
+      "epoch": 5.250852411105699,
+      "grad_norm": 1.2365459203720093,
+      "learning_rate": 0.0002,
+      "loss": 1.0068,
+      "step": 5390
+    },
+    {
+      "epoch": 5.2605942523136875,
+      "grad_norm": 1.3197922706604004,
+      "learning_rate": 0.0002,
+      "loss": 1.0145,
+      "step": 5400
+    },
+    {
+      "epoch": 5.270336093521675,
+      "grad_norm": 1.2395117282867432,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 5410
+    },
+    {
+      "epoch": 5.280077934729664,
+      "grad_norm": 1.1841236352920532,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 5420
+    },
+    {
+      "epoch": 5.289819775937652,
+      "grad_norm": 1.218003749847412,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 5430
+    },
+    {
+      "epoch": 5.29956161714564,
+      "grad_norm": 1.2210947275161743,
+      "learning_rate": 0.0002,
+      "loss": 1.0093,
+      "step": 5440
+    },
+    {
+      "epoch": 5.309303458353629,
+      "grad_norm": 1.266006588935852,
+      "learning_rate": 0.0002,
+      "loss": 0.9619,
+      "step": 5450
+    },
+    {
+      "epoch": 5.319045299561617,
+      "grad_norm": 1.2598075866699219,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 5460
+    },
+    {
+      "epoch": 5.328787140769606,
+      "grad_norm": 1.2410019636154175,
+      "learning_rate": 0.0002,
+      "loss": 1.0929,
+      "step": 5470
+    },
+    {
+      "epoch": 5.338528981977594,
+      "grad_norm": 1.249698519706726,
+      "learning_rate": 0.0002,
+      "loss": 1.058,
+      "step": 5480
+    },
+    {
+      "epoch": 5.348270823185582,
+      "grad_norm": 1.2398173809051514,
+      "learning_rate": 0.0002,
+      "loss": 1.0457,
+      "step": 5490
+    },
+    {
+      "epoch": 5.35801266439357,
+      "grad_norm": 1.2416654825210571,
+      "learning_rate": 0.0002,
+      "loss": 1.0139,
+      "step": 5500
+    },
+    {
+      "epoch": 5.3677545056015585,
+      "grad_norm": 1.398706316947937,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 5510
+    },
+    {
+      "epoch": 5.377496346809547,
+      "grad_norm": 1.3049418926239014,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 5520
+    },
+    {
+      "epoch": 5.387238188017536,
+      "grad_norm": 1.2528893947601318,
+      "learning_rate": 0.0002,
+      "loss": 1.0912,
+      "step": 5530
+    },
+    {
+      "epoch": 5.3969800292255234,
+      "grad_norm": 1.2963255643844604,
+      "learning_rate": 0.0002,
+      "loss": 1.0619,
+      "step": 5540
+    },
+    {
+      "epoch": 5.406721870433512,
+      "grad_norm": 1.494231104850769,
+      "learning_rate": 0.0002,
+      "loss": 1.0194,
+      "step": 5550
+    },
+    {
+      "epoch": 5.416463711641501,
+      "grad_norm": 1.2760992050170898,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 5560
+    },
+    {
+      "epoch": 5.426205552849488,
+      "grad_norm": 1.195292592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.1088,
+      "step": 5570
+    },
+    {
+      "epoch": 5.435947394057477,
+      "grad_norm": 1.6408965587615967,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 5580
+    },
+    {
+      "epoch": 5.4456892352654656,
+      "grad_norm": 1.3092058897018433,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 5590
+    },
+    {
+      "epoch": 5.455431076473453,
+      "grad_norm": 1.2960586547851562,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 5600
+    },
+    {
+      "epoch": 5.465172917681442,
+      "grad_norm": 1.3560487031936646,
+      "learning_rate": 0.0002,
+      "loss": 1.0257,
+      "step": 5610
+    },
+    {
+      "epoch": 5.4749147588894305,
+      "grad_norm": 1.1896311044692993,
+      "learning_rate": 0.0002,
+      "loss": 1.0314,
+      "step": 5620
+    },
+    {
+      "epoch": 5.484656600097418,
+      "grad_norm": 1.3145595788955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0435,
+      "step": 5630
+    },
+    {
+      "epoch": 5.494398441305407,
+      "grad_norm": 1.2207404375076294,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 5640
+    },
+    {
+      "epoch": 5.504140282513395,
+      "grad_norm": 1.266015887260437,
+      "learning_rate": 0.0002,
+      "loss": 1.0823,
+      "step": 5650
+    },
+    {
+      "epoch": 5.513882123721383,
+      "grad_norm": 1.2478289604187012,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 5660
+    },
+    {
+      "epoch": 5.523623964929372,
+      "grad_norm": 1.4851372241973877,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 5670
+    },
+    {
+      "epoch": 5.53336580613736,
+      "grad_norm": 1.4478679895401,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5680
+    },
+    {
+      "epoch": 5.543107647345348,
+      "grad_norm": 1.1079537868499756,
+      "learning_rate": 0.0002,
+      "loss": 1.043,
+      "step": 5690
+    },
+    {
+      "epoch": 5.552849488553337,
+      "grad_norm": 1.4201879501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 5700
+    },
+    {
+      "epoch": 5.562591329761325,
+      "grad_norm": 1.2092000246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.0697,
+      "step": 5710
+    },
+    {
+      "epoch": 5.572333170969313,
+      "grad_norm": 1.4515851736068726,
+      "learning_rate": 0.0002,
+      "loss": 0.9868,
+      "step": 5720
+    },
+    {
+      "epoch": 5.5820750121773015,
+      "grad_norm": 1.3260412216186523,
+      "learning_rate": 0.0002,
+      "loss": 1.1547,
+      "step": 5730
+    },
+    {
+      "epoch": 5.59181685338529,
+      "grad_norm": 1.248191475868225,
+      "learning_rate": 0.0002,
+      "loss": 1.1388,
+      "step": 5740
+    },
+    {
+      "epoch": 5.601558694593278,
+      "grad_norm": 1.2037307024002075,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 5750
+    },
+    {
+      "epoch": 5.611300535801266,
+      "grad_norm": 1.341237187385559,
+      "learning_rate": 0.0002,
+      "loss": 1.1425,
+      "step": 5760
+    },
+    {
+      "epoch": 5.621042377009255,
+      "grad_norm": 1.130115270614624,
+      "learning_rate": 0.0002,
+      "loss": 1.0942,
+      "step": 5770
+    },
+    {
+      "epoch": 5.630784218217243,
+      "grad_norm": 1.3834772109985352,
+      "learning_rate": 0.0002,
+      "loss": 1.1029,
+      "step": 5780
+    },
+    {
+      "epoch": 5.640526059425231,
+      "grad_norm": 1.2586270570755005,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 5790
+    },
+    {
+      "epoch": 5.65026790063322,
+      "grad_norm": 1.3233023881912231,
+      "learning_rate": 0.0002,
+      "loss": 1.0186,
+      "step": 5800
+    },
+    {
+      "epoch": 5.660009741841208,
+      "grad_norm": 1.2711341381072998,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 5810
+    },
+    {
+      "epoch": 5.669751583049196,
+      "grad_norm": 1.3867720365524292,
+      "learning_rate": 0.0002,
+      "loss": 1.0897,
+      "step": 5820
+    },
+    {
+      "epoch": 5.679493424257185,
+      "grad_norm": 1.4783269166946411,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 5830
+    },
+    {
+      "epoch": 5.6892352654651726,
+      "grad_norm": 1.2744768857955933,
+      "learning_rate": 0.0002,
+      "loss": 1.0632,
+      "step": 5840
+    },
+    {
+      "epoch": 5.698977106673161,
+      "grad_norm": 1.3405882120132446,
+      "learning_rate": 0.0002,
+      "loss": 1.1484,
+      "step": 5850
+    },
+    {
+      "epoch": 5.70871894788115,
+      "grad_norm": 1.204300880432129,
+      "learning_rate": 0.0002,
+      "loss": 1.0975,
+      "step": 5860
+    },
+    {
+      "epoch": 5.7184607890891375,
+      "grad_norm": 1.2954572439193726,
+      "learning_rate": 0.0002,
+      "loss": 1.0494,
+      "step": 5870
+    },
+    {
+      "epoch": 5.728202630297126,
+      "grad_norm": 1.5478382110595703,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 5880
+    },
+    {
+      "epoch": 5.737944471505115,
+      "grad_norm": 1.2095842361450195,
+      "learning_rate": 0.0002,
+      "loss": 1.0582,
+      "step": 5890
+    },
+    {
+      "epoch": 5.747686312713103,
+      "grad_norm": 1.0691519975662231,
+      "learning_rate": 0.0002,
+      "loss": 1.1,
+      "step": 5900
+    },
+    {
+      "epoch": 5.757428153921091,
+      "grad_norm": 1.1920677423477173,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 5910
+    },
+    {
+      "epoch": 5.76716999512908,
+      "grad_norm": 1.2051277160644531,
+      "learning_rate": 0.0002,
+      "loss": 1.1746,
+      "step": 5920
+    },
+    {
+      "epoch": 5.776911836337067,
+      "grad_norm": 1.197490930557251,
+      "learning_rate": 0.0002,
+      "loss": 1.1221,
+      "step": 5930
+    },
+    {
+      "epoch": 5.786653677545056,
+      "grad_norm": 1.2003998756408691,
+      "learning_rate": 0.0002,
+      "loss": 1.07,
+      "step": 5940
+    },
+    {
+      "epoch": 5.7963955187530445,
+      "grad_norm": 1.2323646545410156,
+      "learning_rate": 0.0002,
+      "loss": 1.0938,
+      "step": 5950
+    },
+    {
+      "epoch": 5.806137359961033,
+      "grad_norm": 1.2593932151794434,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 5960
+    },
+    {
+      "epoch": 5.815879201169021,
+      "grad_norm": 1.1835976839065552,
+      "learning_rate": 0.0002,
+      "loss": 1.0829,
+      "step": 5970
+    },
+    {
+      "epoch": 5.825621042377009,
+      "grad_norm": 1.4770104885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 5980
+    },
+    {
+      "epoch": 5.835362883584997,
+      "grad_norm": 1.1025809049606323,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 5990
+    },
+    {
+      "epoch": 5.845104724792986,
+      "grad_norm": 1.364588975906372,
+      "learning_rate": 0.0002,
+      "loss": 1.1323,
+      "step": 6000
+    },
+    {
+      "epoch": 5.854846566000974,
+      "grad_norm": 1.2340112924575806,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 6010
+    },
+    {
+      "epoch": 5.864588407208963,
+      "grad_norm": 1.4925711154937744,
+      "learning_rate": 0.0002,
+      "loss": 1.1123,
+      "step": 6020
+    },
+    {
+      "epoch": 5.874330248416951,
+      "grad_norm": 1.3516744375228882,
+      "learning_rate": 0.0002,
+      "loss": 1.12,
+      "step": 6030
+    },
+    {
+      "epoch": 5.884072089624939,
+      "grad_norm": 1.2058138847351074,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 6040
+    },
+    {
+      "epoch": 5.893813930832927,
+      "grad_norm": 1.13870108127594,
+      "learning_rate": 0.0002,
+      "loss": 1.1074,
+      "step": 6050
+    },
+    {
+      "epoch": 5.9035557720409155,
+      "grad_norm": 1.1587319374084473,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 6060
+    },
+    {
+      "epoch": 5.913297613248904,
+      "grad_norm": 1.164481520652771,
+      "learning_rate": 0.0002,
+      "loss": 1.1376,
+      "step": 6070
+    },
+    {
+      "epoch": 5.923039454456893,
+      "grad_norm": 1.2115206718444824,
+      "learning_rate": 0.0002,
+      "loss": 1.1262,
+      "step": 6080
+    },
+    {
+      "epoch": 5.93278129566488,
+      "grad_norm": 1.3201590776443481,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 6090
+    },
+    {
+      "epoch": 5.942523136872869,
+      "grad_norm": 1.287380576133728,
+      "learning_rate": 0.0002,
+      "loss": 1.1288,
+      "step": 6100
+    },
+    {
+      "epoch": 5.952264978080858,
+      "grad_norm": 1.1820166110992432,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 6110
+    },
+    {
+      "epoch": 5.962006819288845,
+      "grad_norm": 1.2550667524337769,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 6120
+    },
+    {
+      "epoch": 5.971748660496834,
+      "grad_norm": 1.3547813892364502,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 6130
+    },
+    {
+      "epoch": 5.9814905017048225,
+      "grad_norm": 1.260842204093933,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 6140
+    },
+    {
+      "epoch": 5.99123234291281,
+      "grad_norm": 1.1643036603927612,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 6150
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.2628161907196045,
+      "eval_runtime": 57.2379,
+      "eval_samples_per_second": 8.858,
+      "eval_steps_per_second": 1.118,
+      "step": 6159
+    },
+    {
+      "epoch": 6.000974184120799,
+      "grad_norm": 0.9384723901748657,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 6160
+    },
+    {
+      "epoch": 6.0107160253287875,
+      "grad_norm": 2.1525821685791016,
+      "learning_rate": 0.0002,
+      "loss": 0.7335,
+      "step": 6170
+    },
+    {
+      "epoch": 6.020457866536775,
+      "grad_norm": 2.0194077491760254,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 6180
+    },
+    {
+      "epoch": 6.030199707744764,
+      "grad_norm": 1.5257816314697266,
+      "learning_rate": 0.0002,
+      "loss": 0.8443,
+      "step": 6190
+    },
+    {
+      "epoch": 6.039941548952752,
+      "grad_norm": 1.5432662963867188,
+      "learning_rate": 0.0002,
+      "loss": 0.7543,
+      "step": 6200
+    },
+    {
+      "epoch": 6.04968339016074,
+      "grad_norm": 1.6874405145645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8104,
+      "step": 6210
+    },
+    {
+      "epoch": 6.059425231368729,
+      "grad_norm": 1.7346407175064087,
+      "learning_rate": 0.0002,
+      "loss": 0.8395,
+      "step": 6220
+    },
+    {
+      "epoch": 6.069167072576717,
+      "grad_norm": 1.5320781469345093,
+      "learning_rate": 0.0002,
+      "loss": 0.8027,
+      "step": 6230
+    },
+    {
+      "epoch": 6.078908913784705,
+      "grad_norm": 1.4106669425964355,
+      "learning_rate": 0.0002,
+      "loss": 0.7488,
+      "step": 6240
+    },
+    {
+      "epoch": 6.088650754992694,
+      "grad_norm": 1.5568628311157227,
+      "learning_rate": 0.0002,
+      "loss": 0.812,
+      "step": 6250
+    },
+    {
+      "epoch": 6.098392596200682,
+      "grad_norm": 1.6155978441238403,
+      "learning_rate": 0.0002,
+      "loss": 0.8055,
+      "step": 6260
+    },
+    {
+      "epoch": 6.10813443740867,
+      "grad_norm": 1.4820445775985718,
+      "learning_rate": 0.0002,
+      "loss": 0.8225,
+      "step": 6270
+    },
+    {
+      "epoch": 6.1178762786166585,
+      "grad_norm": 1.6163820028305054,
+      "learning_rate": 0.0002,
+      "loss": 0.8599,
+      "step": 6280
+    },
+    {
+      "epoch": 6.127618119824647,
+      "grad_norm": 1.8396387100219727,
+      "learning_rate": 0.0002,
+      "loss": 0.853,
+      "step": 6290
+    },
+    {
+      "epoch": 6.137359961032635,
+      "grad_norm": 1.7181230783462524,
+      "learning_rate": 0.0002,
+      "loss": 0.7768,
+      "step": 6300
+    },
+    {
+      "epoch": 6.147101802240623,
+      "grad_norm": 1.6568509340286255,
+      "learning_rate": 0.0002,
+      "loss": 0.8116,
+      "step": 6310
+    },
+    {
+      "epoch": 6.156843643448612,
+      "grad_norm": 1.3481947183609009,
+      "learning_rate": 0.0002,
+      "loss": 0.8525,
+      "step": 6320
+    },
+    {
+      "epoch": 6.1665854846566,
+      "grad_norm": 1.5788342952728271,
+      "learning_rate": 0.0002,
+      "loss": 0.762,
+      "step": 6330
+    },
+    {
+      "epoch": 6.176327325864588,
+      "grad_norm": 1.5067620277404785,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 6340
+    },
+    {
+      "epoch": 6.186069167072577,
+      "grad_norm": 1.8198208808898926,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 6350
+    },
+    {
+      "epoch": 6.195811008280565,
+      "grad_norm": 1.4012749195098877,
+      "learning_rate": 0.0002,
+      "loss": 0.7867,
+      "step": 6360
+    },
+    {
+      "epoch": 6.205552849488553,
+      "grad_norm": 1.759798288345337,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 6370
+    },
+    {
+      "epoch": 6.215294690696542,
+      "grad_norm": 1.468922734260559,
+      "learning_rate": 0.0002,
+      "loss": 0.7811,
+      "step": 6380
+    },
+    {
+      "epoch": 6.2250365319045295,
+      "grad_norm": 1.3706471920013428,
+      "learning_rate": 0.0002,
+      "loss": 0.8356,
+      "step": 6390
+    },
+    {
+      "epoch": 6.234778373112518,
+      "grad_norm": 1.6397383213043213,
+      "learning_rate": 0.0002,
+      "loss": 0.8096,
+      "step": 6400
+    },
+    {
+      "epoch": 6.244520214320507,
+      "grad_norm": 1.5614187717437744,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 6410
+    },
+    {
+      "epoch": 6.2542620555284945,
+      "grad_norm": 1.7118678092956543,
+      "learning_rate": 0.0002,
+      "loss": 0.8533,
+      "step": 6420
+    },
+    {
+      "epoch": 6.264003896736483,
+      "grad_norm": 1.4041547775268555,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 6430
+    },
+    {
+      "epoch": 6.273745737944472,
+      "grad_norm": 1.7653605937957764,
+      "learning_rate": 0.0002,
+      "loss": 0.879,
+      "step": 6440
+    },
+    {
+      "epoch": 6.28348757915246,
+      "grad_norm": 2.6219191551208496,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 6450
+    },
+    {
+      "epoch": 6.293229420360448,
+      "grad_norm": 1.4757837057113647,
+      "learning_rate": 0.0002,
+      "loss": 0.8896,
+      "step": 6460
+    },
+    {
+      "epoch": 6.302971261568437,
+      "grad_norm": 1.715598225593567,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 6470
+    },
+    {
+      "epoch": 6.312713102776424,
+      "grad_norm": 1.376216173171997,
+      "learning_rate": 0.0002,
+      "loss": 0.8526,
+      "step": 6480
+    },
+    {
+      "epoch": 6.322454943984413,
+      "grad_norm": 1.7119828462600708,
+      "learning_rate": 0.0002,
+      "loss": 0.8742,
+      "step": 6490
+    },
+    {
+      "epoch": 6.3321967851924015,
+      "grad_norm": 1.4304355382919312,
+      "learning_rate": 0.0002,
+      "loss": 0.7988,
+      "step": 6500
+    },
+    {
+      "epoch": 6.34193862640039,
+      "grad_norm": 1.4889872074127197,
+      "learning_rate": 0.0002,
+      "loss": 0.8539,
+      "step": 6510
+    },
+    {
+      "epoch": 6.351680467608378,
+      "grad_norm": 1.370373010635376,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 6520
+    },
+    {
+      "epoch": 6.361422308816366,
+      "grad_norm": 1.7697709798812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 6530
+    },
+    {
+      "epoch": 6.371164150024355,
+      "grad_norm": 1.495297908782959,
+      "learning_rate": 0.0002,
+      "loss": 0.9421,
+      "step": 6540
+    },
+    {
+      "epoch": 6.380905991232343,
+      "grad_norm": 1.7251347303390503,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 6550
+    },
+    {
+      "epoch": 6.390647832440331,
+      "grad_norm": 1.6909505128860474,
+      "learning_rate": 0.0002,
+      "loss": 0.9327,
+      "step": 6560
+    },
+    {
+      "epoch": 6.40038967364832,
+      "grad_norm": 1.4369314908981323,
+      "learning_rate": 0.0002,
+      "loss": 0.837,
+      "step": 6570
+    },
+    {
+      "epoch": 6.410131514856308,
+      "grad_norm": 1.7803739309310913,
+      "learning_rate": 0.0002,
+      "loss": 0.8572,
+      "step": 6580
+    },
+    {
+      "epoch": 6.419873356064296,
+      "grad_norm": 1.6107097864151,
+      "learning_rate": 0.0002,
+      "loss": 0.9024,
+      "step": 6590
+    },
+    {
+      "epoch": 6.429615197272285,
+      "grad_norm": 1.6151643991470337,
+      "learning_rate": 0.0002,
+      "loss": 0.8469,
+      "step": 6600
+    },
+    {
+      "epoch": 6.4393570384802725,
+      "grad_norm": 1.7159833908081055,
+      "learning_rate": 0.0002,
+      "loss": 0.8791,
+      "step": 6610
+    },
+    {
+      "epoch": 6.449098879688261,
+      "grad_norm": 1.4366064071655273,
+      "learning_rate": 0.0002,
+      "loss": 0.9249,
+      "step": 6620
+    },
+    {
+      "epoch": 6.45884072089625,
+      "grad_norm": 1.6050453186035156,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 6630
+    },
+    {
+      "epoch": 6.468582562104237,
+      "grad_norm": 1.6296740770339966,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 6640
+    },
+    {
+      "epoch": 6.478324403312226,
+      "grad_norm": 1.6181174516677856,
+      "learning_rate": 0.0002,
+      "loss": 0.9228,
+      "step": 6650
+    },
+    {
+      "epoch": 6.488066244520215,
+      "grad_norm": 1.5452176332473755,
+      "learning_rate": 0.0002,
+      "loss": 0.9139,
+      "step": 6660
+    },
+    {
+      "epoch": 6.497808085728202,
+      "grad_norm": 1.3919731378555298,
+      "learning_rate": 0.0002,
+      "loss": 0.9022,
+      "step": 6670
+    },
+    {
+      "epoch": 6.507549926936191,
+      "grad_norm": 1.6456257104873657,
+      "learning_rate": 0.0002,
+      "loss": 0.9046,
+      "step": 6680
+    },
+    {
+      "epoch": 6.5172917681441795,
+      "grad_norm": 1.4147369861602783,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 6690
+    },
+    {
+      "epoch": 6.527033609352167,
+      "grad_norm": 1.7005025148391724,
+      "learning_rate": 0.0002,
+      "loss": 0.8361,
+      "step": 6700
+    },
+    {
+      "epoch": 6.536775450560156,
+      "grad_norm": 1.6032357215881348,
+      "learning_rate": 0.0002,
+      "loss": 0.8738,
+      "step": 6710
+    },
+    {
+      "epoch": 6.5465172917681445,
+      "grad_norm": 1.3454229831695557,
+      "learning_rate": 0.0002,
+      "loss": 0.9796,
+      "step": 6720
+    },
+    {
+      "epoch": 6.556259132976132,
+      "grad_norm": 1.6961418390274048,
+      "learning_rate": 0.0002,
+      "loss": 0.8573,
+      "step": 6730
+    },
+    {
+      "epoch": 6.566000974184121,
+      "grad_norm": 1.78407883644104,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 6740
+    },
+    {
+      "epoch": 6.575742815392109,
+      "grad_norm": 1.6817889213562012,
+      "learning_rate": 0.0002,
+      "loss": 0.8941,
+      "step": 6750
+    },
+    {
+      "epoch": 6.585484656600097,
+      "grad_norm": 1.7894943952560425,
+      "learning_rate": 0.0002,
+      "loss": 0.8765,
+      "step": 6760
+    },
+    {
+      "epoch": 6.595226497808086,
+      "grad_norm": 1.6404837369918823,
+      "learning_rate": 0.0002,
+      "loss": 0.8607,
+      "step": 6770
+    },
+    {
+      "epoch": 6.604968339016074,
+      "grad_norm": 1.5849255323410034,
+      "learning_rate": 0.0002,
+      "loss": 0.8573,
+      "step": 6780
+    },
+    {
+      "epoch": 6.614710180224062,
+      "grad_norm": 1.5993813276290894,
+      "learning_rate": 0.0002,
+      "loss": 0.9575,
+      "step": 6790
+    },
+    {
+      "epoch": 6.624452021432051,
+      "grad_norm": 1.2834863662719727,
+      "learning_rate": 0.0002,
+      "loss": 0.8922,
+      "step": 6800
+    },
+    {
+      "epoch": 6.634193862640039,
+      "grad_norm": 1.7215641736984253,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 6810
+    },
+    {
+      "epoch": 6.643935703848027,
+      "grad_norm": 1.7588146924972534,
+      "learning_rate": 0.0002,
+      "loss": 0.9292,
+      "step": 6820
+    },
+    {
+      "epoch": 6.6536775450560155,
+      "grad_norm": 1.7956023216247559,
+      "learning_rate": 0.0002,
+      "loss": 0.8634,
+      "step": 6830
+    },
+    {
+      "epoch": 6.663419386264004,
+      "grad_norm": 1.5115351676940918,
+      "learning_rate": 0.0002,
+      "loss": 0.8108,
+      "step": 6840
+    },
+    {
+      "epoch": 6.673161227471992,
+      "grad_norm": 1.5660319328308105,
+      "learning_rate": 0.0002,
+      "loss": 0.9329,
+      "step": 6850
+    },
+    {
+      "epoch": 6.68290306867998,
+      "grad_norm": 1.4323679208755493,
+      "learning_rate": 0.0002,
+      "loss": 0.9877,
+      "step": 6860
+    },
+    {
+      "epoch": 6.692644909887969,
+      "grad_norm": 1.662089467048645,
+      "learning_rate": 0.0002,
+      "loss": 0.8732,
+      "step": 6870
+    },
+    {
+      "epoch": 6.702386751095958,
+      "grad_norm": 1.7854869365692139,
+      "learning_rate": 0.0002,
+      "loss": 0.87,
+      "step": 6880
+    },
+    {
+      "epoch": 6.712128592303945,
+      "grad_norm": 1.5491222143173218,
+      "learning_rate": 0.0002,
+      "loss": 0.9105,
+      "step": 6890
+    },
+    {
+      "epoch": 6.721870433511934,
+      "grad_norm": 1.5946987867355347,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 6900
+    },
+    {
+      "epoch": 6.731612274719922,
+      "grad_norm": 1.6195964813232422,
+      "learning_rate": 0.0002,
+      "loss": 0.9391,
+      "step": 6910
+    },
+    {
+      "epoch": 6.74135411592791,
+      "grad_norm": 1.6366901397705078,
+      "learning_rate": 0.0002,
+      "loss": 0.8947,
+      "step": 6920
+    },
+    {
+      "epoch": 6.751095957135899,
+      "grad_norm": 1.5080382823944092,
+      "learning_rate": 0.0002,
+      "loss": 0.8695,
+      "step": 6930
+    },
+    {
+      "epoch": 6.760837798343887,
+      "grad_norm": 1.742353916168213,
+      "learning_rate": 0.0002,
+      "loss": 0.9124,
+      "step": 6940
+    },
+    {
+      "epoch": 6.770579639551875,
+      "grad_norm": 1.690251111984253,
+      "learning_rate": 0.0002,
+      "loss": 0.9118,
+      "step": 6950
+    },
+    {
+      "epoch": 6.780321480759864,
+      "grad_norm": 1.7103357315063477,
+      "learning_rate": 0.0002,
+      "loss": 0.9039,
+      "step": 6960
+    },
+    {
+      "epoch": 6.7900633219678515,
+      "grad_norm": 1.6630914211273193,
+      "learning_rate": 0.0002,
+      "loss": 0.869,
+      "step": 6970
+    },
+    {
+      "epoch": 6.79980516317584,
+      "grad_norm": 1.423768162727356,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 6980
+    },
+    {
+      "epoch": 6.809547004383829,
+      "grad_norm": 1.7844693660736084,
+      "learning_rate": 0.0002,
+      "loss": 0.9397,
+      "step": 6990
+    },
+    {
+      "epoch": 6.819288845591817,
+      "grad_norm": 1.545282006263733,
+      "learning_rate": 0.0002,
+      "loss": 0.8889,
+      "step": 7000
+    },
+    {
+      "epoch": 6.829030686799805,
+      "grad_norm": 1.4340319633483887,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 7010
+    },
+    {
+      "epoch": 6.838772528007794,
+      "grad_norm": 1.5981626510620117,
+      "learning_rate": 0.0002,
+      "loss": 0.9486,
+      "step": 7020
+    },
+    {
+      "epoch": 6.848514369215782,
+      "grad_norm": 1.5205026865005493,
+      "learning_rate": 0.0002,
+      "loss": 0.9062,
+      "step": 7030
+    },
+    {
+      "epoch": 6.85825621042377,
+      "grad_norm": 1.6999989748001099,
+      "learning_rate": 0.0002,
+      "loss": 0.9245,
+      "step": 7040
+    },
+    {
+      "epoch": 6.8679980516317585,
+      "grad_norm": 1.6392347812652588,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 7050
+    },
+    {
+      "epoch": 6.877739892839747,
+      "grad_norm": 1.637308955192566,
+      "learning_rate": 0.0002,
+      "loss": 0.9275,
+      "step": 7060
+    },
+    {
+      "epoch": 6.887481734047735,
+      "grad_norm": 1.671341896057129,
+      "learning_rate": 0.0002,
+      "loss": 0.9672,
+      "step": 7070
+    },
+    {
+      "epoch": 6.897223575255723,
+      "grad_norm": 1.4437555074691772,
+      "learning_rate": 0.0002,
+      "loss": 0.9726,
+      "step": 7080
+    },
+    {
+      "epoch": 6.906965416463712,
+      "grad_norm": 1.4251935482025146,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 7090
+    },
+    {
+      "epoch": 6.9167072576717,
+      "grad_norm": 1.5106734037399292,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 7100
+    },
+    {
+      "epoch": 6.926449098879688,
+      "grad_norm": 1.670742154121399,
+      "learning_rate": 0.0002,
+      "loss": 0.939,
+      "step": 7110
+    },
+    {
+      "epoch": 6.936190940087677,
+      "grad_norm": 1.4353723526000977,
+      "learning_rate": 0.0002,
+      "loss": 0.8818,
+      "step": 7120
+    },
+    {
+      "epoch": 6.945932781295665,
+      "grad_norm": 1.9437772035598755,
+      "learning_rate": 0.0002,
+      "loss": 0.9354,
+      "step": 7130
+    },
+    {
+      "epoch": 6.955674622503653,
+      "grad_norm": 1.4922038316726685,
+      "learning_rate": 0.0002,
+      "loss": 0.9623,
+      "step": 7140
+    },
+    {
+      "epoch": 6.965416463711642,
+      "grad_norm": 1.489193081855774,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 7150
+    },
+    {
+      "epoch": 6.9751583049196295,
+      "grad_norm": 1.529490351676941,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 7160
+    },
+    {
+      "epoch": 6.984900146127618,
+      "grad_norm": 1.7370105981826782,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 7170
+    },
+    {
+      "epoch": 6.994641987335607,
+      "grad_norm": 1.5639604330062866,
+      "learning_rate": 0.0002,
+      "loss": 0.921,
+      "step": 7180
+    },
+    {
+      "epoch": 6.9995129079396,
+      "eval_loss": 2.521758794784546,
+      "eval_runtime": 56.1587,
+      "eval_samples_per_second": 9.028,
+      "eval_steps_per_second": 1.14,
+      "step": 7185
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "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": 3.690741942111437e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-7185/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/README.md b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_model.safetensors b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5433538c92f49a9f359ba0c5182ab149dcf1a196
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e2ff42e571db69511d0848cefcaf9b1f9e59bf1c3e3ff6a29e447e500c46c59
+size 143153376
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/optimizer.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0e8de090c68300eab35c5d8fcfd2ddc0f40b2822
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd0e8db97eb715d41d3018b5c2c2d671cb3796d3b95a0cbb6c266c4c8af5439b
+size 72886650
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/rng_state.pth b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7dd8bf7456fad9edd68113143d4a1cd8d1cc947e
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7432f83a4c6152fd92ed8ca152d25a2c7c587d2430fe61817434dc641419d9
+size 14244
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/scheduler.pt b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..82d33307f21179c52d408ea20eeb125b85b01b25
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a123ed91b99863f6a4bfffa749f6e9055a4cad42fec0e03d6a304d1cc972c68f
+size 1064
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/trainer_state.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..99932528bdebbe9c06f4d5ebbaa98b1ec80f8833
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/trainer_state.json
@@ -0,0 +1,5837 @@
+{
+  "best_metric": 1.8103164434432983,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026",
+  "epoch": 7.996103263516805,
+  "eval_steps": 10,
+  "global_step": 8208,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00974184120798831,
+      "grad_norm": 0.6537588834762573,
+      "learning_rate": 0.0002,
+      "loss": 3.0782,
+      "step": 10
+    },
+    {
+      "epoch": 0.01948368241597662,
+      "grad_norm": 0.5270306468009949,
+      "learning_rate": 0.0002,
+      "loss": 2.4644,
+      "step": 20
+    },
+    {
+      "epoch": 0.029225523623964928,
+      "grad_norm": 0.6826501488685608,
+      "learning_rate": 0.0002,
+      "loss": 2.2589,
+      "step": 30
+    },
+    {
+      "epoch": 0.03896736483195324,
+      "grad_norm": 0.5061377286911011,
+      "learning_rate": 0.0002,
+      "loss": 2.0141,
+      "step": 40
+    },
+    {
+      "epoch": 0.04870920603994155,
+      "grad_norm": 0.4300410747528076,
+      "learning_rate": 0.0002,
+      "loss": 1.9458,
+      "step": 50
+    },
+    {
+      "epoch": 0.058451047247929856,
+      "grad_norm": 0.5063319802284241,
+      "learning_rate": 0.0002,
+      "loss": 1.983,
+      "step": 60
+    },
+    {
+      "epoch": 0.06819288845591817,
+      "grad_norm": 0.49310117959976196,
+      "learning_rate": 0.0002,
+      "loss": 1.9799,
+      "step": 70
+    },
+    {
+      "epoch": 0.07793472966390648,
+      "grad_norm": 0.4676004648208618,
+      "learning_rate": 0.0002,
+      "loss": 1.9277,
+      "step": 80
+    },
+    {
+      "epoch": 0.08767657087189479,
+      "grad_norm": 0.41647228598594666,
+      "learning_rate": 0.0002,
+      "loss": 1.9147,
+      "step": 90
+    },
+    {
+      "epoch": 0.0974184120798831,
+      "grad_norm": 0.40217313170433044,
+      "learning_rate": 0.0002,
+      "loss": 1.8894,
+      "step": 100
+    },
+    {
+      "epoch": 0.1071602532878714,
+      "grad_norm": 0.4123637080192566,
+      "learning_rate": 0.0002,
+      "loss": 1.9099,
+      "step": 110
+    },
+    {
+      "epoch": 0.11690209449585971,
+      "grad_norm": 0.37414297461509705,
+      "learning_rate": 0.0002,
+      "loss": 1.8471,
+      "step": 120
+    },
+    {
+      "epoch": 0.12664393570384802,
+      "grad_norm": 0.3179326355457306,
+      "learning_rate": 0.0002,
+      "loss": 1.894,
+      "step": 130
+    },
+    {
+      "epoch": 0.13638577691183634,
+      "grad_norm": 0.3548192083835602,
+      "learning_rate": 0.0002,
+      "loss": 1.847,
+      "step": 140
+    },
+    {
+      "epoch": 0.14612761811982464,
+      "grad_norm": 0.3273540139198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 150
+    },
+    {
+      "epoch": 0.15586945932781296,
+      "grad_norm": 0.36500975489616394,
+      "learning_rate": 0.0002,
+      "loss": 1.8496,
+      "step": 160
+    },
+    {
+      "epoch": 0.16561130053580125,
+      "grad_norm": 0.4106619656085968,
+      "learning_rate": 0.0002,
+      "loss": 1.8473,
+      "step": 170
+    },
+    {
+      "epoch": 0.17535314174378958,
+      "grad_norm": 0.41361644864082336,
+      "learning_rate": 0.0002,
+      "loss": 1.841,
+      "step": 180
+    },
+    {
+      "epoch": 0.1850949829517779,
+      "grad_norm": 0.3608580231666565,
+      "learning_rate": 0.0002,
+      "loss": 1.8879,
+      "step": 190
+    },
+    {
+      "epoch": 0.1948368241597662,
+      "grad_norm": 0.4291760325431824,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 200
+    },
+    {
+      "epoch": 0.20457866536775451,
+      "grad_norm": 0.344184011220932,
+      "learning_rate": 0.0002,
+      "loss": 1.8437,
+      "step": 210
+    },
+    {
+      "epoch": 0.2143205065757428,
+      "grad_norm": 0.3834705650806427,
+      "learning_rate": 0.0002,
+      "loss": 1.8779,
+      "step": 220
+    },
+    {
+      "epoch": 0.22406234778373113,
+      "grad_norm": 0.3738210201263428,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 230
+    },
+    {
+      "epoch": 0.23380418899171942,
+      "grad_norm": 0.4306780695915222,
+      "learning_rate": 0.0002,
+      "loss": 1.824,
+      "step": 240
+    },
+    {
+      "epoch": 0.24354603019970775,
+      "grad_norm": 0.5066465139389038,
+      "learning_rate": 0.0002,
+      "loss": 1.8519,
+      "step": 250
+    },
+    {
+      "epoch": 0.25328787140769604,
+      "grad_norm": 0.34227681159973145,
+      "learning_rate": 0.0002,
+      "loss": 1.7402,
+      "step": 260
+    },
+    {
+      "epoch": 0.26302971261568436,
+      "grad_norm": 0.3346865475177765,
+      "learning_rate": 0.0002,
+      "loss": 1.8614,
+      "step": 270
+    },
+    {
+      "epoch": 0.2727715538236727,
+      "grad_norm": 0.3639362156391144,
+      "learning_rate": 0.0002,
+      "loss": 1.8502,
+      "step": 280
+    },
+    {
+      "epoch": 0.282513395031661,
+      "grad_norm": 0.33223700523376465,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 290
+    },
+    {
+      "epoch": 0.2922552362396493,
+      "grad_norm": 0.35176315903663635,
+      "learning_rate": 0.0002,
+      "loss": 1.821,
+      "step": 300
+    },
+    {
+      "epoch": 0.3019970774476376,
+      "grad_norm": 0.3581472635269165,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 310
+    },
+    {
+      "epoch": 0.3117389186556259,
+      "grad_norm": 0.35943421721458435,
+      "learning_rate": 0.0002,
+      "loss": 1.8262,
+      "step": 320
+    },
+    {
+      "epoch": 0.32148075986361424,
+      "grad_norm": 0.322051078081131,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 330
+    },
+    {
+      "epoch": 0.3312226010716025,
+      "grad_norm": 0.33904823660850525,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 340
+    },
+    {
+      "epoch": 0.34096444227959083,
+      "grad_norm": 0.39162731170654297,
+      "learning_rate": 0.0002,
+      "loss": 1.9159,
+      "step": 350
+    },
+    {
+      "epoch": 0.35070628348757915,
+      "grad_norm": 0.330624520778656,
+      "learning_rate": 0.0002,
+      "loss": 1.7966,
+      "step": 360
+    },
+    {
+      "epoch": 0.3604481246955675,
+      "grad_norm": 0.3793248236179352,
+      "learning_rate": 0.0002,
+      "loss": 1.8646,
+      "step": 370
+    },
+    {
+      "epoch": 0.3701899659035558,
+      "grad_norm": 0.3347395658493042,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 380
+    },
+    {
+      "epoch": 0.37993180711154406,
+      "grad_norm": 0.30527254939079285,
+      "learning_rate": 0.0002,
+      "loss": 1.9244,
+      "step": 390
+    },
+    {
+      "epoch": 0.3896736483195324,
+      "grad_norm": 0.3081390857696533,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 400
+    },
+    {
+      "epoch": 0.3994154895275207,
+      "grad_norm": 0.3742620050907135,
+      "learning_rate": 0.0002,
+      "loss": 1.8968,
+      "step": 410
+    },
+    {
+      "epoch": 0.40915733073550903,
+      "grad_norm": 0.4080568253993988,
+      "learning_rate": 0.0002,
+      "loss": 1.8095,
+      "step": 420
+    },
+    {
+      "epoch": 0.4188991719434973,
+      "grad_norm": 0.38034746050834656,
+      "learning_rate": 0.0002,
+      "loss": 1.8555,
+      "step": 430
+    },
+    {
+      "epoch": 0.4286410131514856,
+      "grad_norm": 0.34893402457237244,
+      "learning_rate": 0.0002,
+      "loss": 1.8494,
+      "step": 440
+    },
+    {
+      "epoch": 0.43838285435947394,
+      "grad_norm": 0.33285608887672424,
+      "learning_rate": 0.0002,
+      "loss": 1.8481,
+      "step": 450
+    },
+    {
+      "epoch": 0.44812469556746226,
+      "grad_norm": 0.4110095798969269,
+      "learning_rate": 0.0002,
+      "loss": 1.8466,
+      "step": 460
+    },
+    {
+      "epoch": 0.4578665367754506,
+      "grad_norm": 0.3658817410469055,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 470
+    },
+    {
+      "epoch": 0.46760837798343885,
+      "grad_norm": 0.31350770592689514,
+      "learning_rate": 0.0002,
+      "loss": 1.7589,
+      "step": 480
+    },
+    {
+      "epoch": 0.47735021919142717,
+      "grad_norm": 0.38827991485595703,
+      "learning_rate": 0.0002,
+      "loss": 1.7839,
+      "step": 490
+    },
+    {
+      "epoch": 0.4870920603994155,
+      "grad_norm": 0.3792393207550049,
+      "learning_rate": 0.0002,
+      "loss": 1.8224,
+      "step": 500
+    },
+    {
+      "epoch": 0.4968339016074038,
+      "grad_norm": 0.3004095256328583,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 510
+    },
+    {
+      "epoch": 0.5065757428153921,
+      "grad_norm": 0.3200063407421112,
+      "learning_rate": 0.0002,
+      "loss": 1.6899,
+      "step": 520
+    },
+    {
+      "epoch": 0.5163175840233805,
+      "grad_norm": 0.3206128478050232,
+      "learning_rate": 0.0002,
+      "loss": 1.8205,
+      "step": 530
+    },
+    {
+      "epoch": 0.5260594252313687,
+      "grad_norm": 0.30258631706237793,
+      "learning_rate": 0.0002,
+      "loss": 1.7725,
+      "step": 540
+    },
+    {
+      "epoch": 0.535801266439357,
+      "grad_norm": 0.28210392594337463,
+      "learning_rate": 0.0002,
+      "loss": 1.7791,
+      "step": 550
+    },
+    {
+      "epoch": 0.5455431076473454,
+      "grad_norm": 0.34854066371917725,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 560
+    },
+    {
+      "epoch": 0.5552849488553336,
+      "grad_norm": 0.31689873337745667,
+      "learning_rate": 0.0002,
+      "loss": 1.8331,
+      "step": 570
+    },
+    {
+      "epoch": 0.565026790063322,
+      "grad_norm": 0.31253790855407715,
+      "learning_rate": 0.0002,
+      "loss": 1.8311,
+      "step": 580
+    },
+    {
+      "epoch": 0.5747686312713103,
+      "grad_norm": 0.3229721188545227,
+      "learning_rate": 0.0002,
+      "loss": 1.7035,
+      "step": 590
+    },
+    {
+      "epoch": 0.5845104724792985,
+      "grad_norm": 0.3723772466182709,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 600
+    },
+    {
+      "epoch": 0.5942523136872869,
+      "grad_norm": 0.345798522233963,
+      "learning_rate": 0.0002,
+      "loss": 1.8357,
+      "step": 610
+    },
+    {
+      "epoch": 0.6039941548952752,
+      "grad_norm": 0.3440598249435425,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 620
+    },
+    {
+      "epoch": 0.6137359961032636,
+      "grad_norm": 0.3406416177749634,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 630
+    },
+    {
+      "epoch": 0.6234778373112518,
+      "grad_norm": 0.3218357264995575,
+      "learning_rate": 0.0002,
+      "loss": 1.8103,
+      "step": 640
+    },
+    {
+      "epoch": 0.6332196785192401,
+      "grad_norm": 0.45319172739982605,
+      "learning_rate": 0.0002,
+      "loss": 1.7457,
+      "step": 650
+    },
+    {
+      "epoch": 0.6429615197272285,
+      "grad_norm": 0.2787110507488251,
+      "learning_rate": 0.0002,
+      "loss": 1.8674,
+      "step": 660
+    },
+    {
+      "epoch": 0.6527033609352167,
+      "grad_norm": 0.3064707815647125,
+      "learning_rate": 0.0002,
+      "loss": 1.8426,
+      "step": 670
+    },
+    {
+      "epoch": 0.662445202143205,
+      "grad_norm": 0.2940629720687866,
+      "learning_rate": 0.0002,
+      "loss": 1.846,
+      "step": 680
+    },
+    {
+      "epoch": 0.6721870433511934,
+      "grad_norm": 0.31695225834846497,
+      "learning_rate": 0.0002,
+      "loss": 1.7865,
+      "step": 690
+    },
+    {
+      "epoch": 0.6819288845591817,
+      "grad_norm": 0.29589611291885376,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 700
+    },
+    {
+      "epoch": 0.69167072576717,
+      "grad_norm": 0.3062121570110321,
+      "learning_rate": 0.0002,
+      "loss": 1.7828,
+      "step": 710
+    },
+    {
+      "epoch": 0.7014125669751583,
+      "grad_norm": 0.3315656781196594,
+      "learning_rate": 0.0002,
+      "loss": 1.8108,
+      "step": 720
+    },
+    {
+      "epoch": 0.7111544081831466,
+      "grad_norm": 0.30353930592536926,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 730
+    },
+    {
+      "epoch": 0.720896249391135,
+      "grad_norm": 0.28360483050346375,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 740
+    },
+    {
+      "epoch": 0.7306380905991232,
+      "grad_norm": 0.3362562656402588,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 750
+    },
+    {
+      "epoch": 0.7403799318071116,
+      "grad_norm": 0.40434667468070984,
+      "learning_rate": 0.0002,
+      "loss": 1.7607,
+      "step": 760
+    },
+    {
+      "epoch": 0.7501217730150999,
+      "grad_norm": 0.2930425703525543,
+      "learning_rate": 0.0002,
+      "loss": 1.8374,
+      "step": 770
+    },
+    {
+      "epoch": 0.7598636142230881,
+      "grad_norm": 0.30177003145217896,
+      "learning_rate": 0.0002,
+      "loss": 1.8216,
+      "step": 780
+    },
+    {
+      "epoch": 0.7696054554310765,
+      "grad_norm": 0.2784474790096283,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 790
+    },
+    {
+      "epoch": 0.7793472966390648,
+      "grad_norm": 0.35849854350090027,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 800
+    },
+    {
+      "epoch": 0.7890891378470531,
+      "grad_norm": 0.27329114079475403,
+      "learning_rate": 0.0002,
+      "loss": 1.7896,
+      "step": 810
+    },
+    {
+      "epoch": 0.7988309790550414,
+      "grad_norm": 0.33331671357154846,
+      "learning_rate": 0.0002,
+      "loss": 1.7633,
+      "step": 820
+    },
+    {
+      "epoch": 0.8085728202630297,
+      "grad_norm": 0.28727295994758606,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 830
+    },
+    {
+      "epoch": 0.8183146614710181,
+      "grad_norm": 0.31391268968582153,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 840
+    },
+    {
+      "epoch": 0.8280565026790063,
+      "grad_norm": 0.3303709030151367,
+      "learning_rate": 0.0002,
+      "loss": 1.793,
+      "step": 850
+    },
+    {
+      "epoch": 0.8377983438869946,
+      "grad_norm": 0.33772537112236023,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 860
+    },
+    {
+      "epoch": 0.847540185094983,
+      "grad_norm": 0.32876333594322205,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 870
+    },
+    {
+      "epoch": 0.8572820263029712,
+      "grad_norm": 0.28444716334342957,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 880
+    },
+    {
+      "epoch": 0.8670238675109596,
+      "grad_norm": 0.3070019483566284,
+      "learning_rate": 0.0002,
+      "loss": 1.7077,
+      "step": 890
+    },
+    {
+      "epoch": 0.8767657087189479,
+      "grad_norm": 0.29484760761260986,
+      "learning_rate": 0.0002,
+      "loss": 1.7699,
+      "step": 900
+    },
+    {
+      "epoch": 0.8865075499269361,
+      "grad_norm": 0.32373034954071045,
+      "learning_rate": 0.0002,
+      "loss": 1.7211,
+      "step": 910
+    },
+    {
+      "epoch": 0.8962493911349245,
+      "grad_norm": 0.3229396939277649,
+      "learning_rate": 0.0002,
+      "loss": 1.7799,
+      "step": 920
+    },
+    {
+      "epoch": 0.9059912323429128,
+      "grad_norm": 0.33151453733444214,
+      "learning_rate": 0.0002,
+      "loss": 1.8226,
+      "step": 930
+    },
+    {
+      "epoch": 0.9157330735509012,
+      "grad_norm": 0.32037460803985596,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 940
+    },
+    {
+      "epoch": 0.9254749147588894,
+      "grad_norm": 0.31283533573150635,
+      "learning_rate": 0.0002,
+      "loss": 1.822,
+      "step": 950
+    },
+    {
+      "epoch": 0.9352167559668777,
+      "grad_norm": 0.27984118461608887,
+      "learning_rate": 0.0002,
+      "loss": 1.8233,
+      "step": 960
+    },
+    {
+      "epoch": 0.9449585971748661,
+      "grad_norm": 0.316500186920166,
+      "learning_rate": 0.0002,
+      "loss": 1.7755,
+      "step": 970
+    },
+    {
+      "epoch": 0.9547004383828543,
+      "grad_norm": 0.33708682656288147,
+      "learning_rate": 0.0002,
+      "loss": 1.8032,
+      "step": 980
+    },
+    {
+      "epoch": 0.9644422795908427,
+      "grad_norm": 0.31026017665863037,
+      "learning_rate": 0.0002,
+      "loss": 1.8863,
+      "step": 990
+    },
+    {
+      "epoch": 0.974184120798831,
+      "grad_norm": 0.30874672532081604,
+      "learning_rate": 0.0002,
+      "loss": 1.8458,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9839259620068193,
+      "grad_norm": 0.3257741630077362,
+      "learning_rate": 0.0002,
+      "loss": 1.7975,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9936678032148076,
+      "grad_norm": 0.2865653932094574,
+      "learning_rate": 0.0002,
+      "loss": 1.7936,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9995129079396006,
+      "eval_loss": 1.8103164434432983,
+      "eval_runtime": 56.3917,
+      "eval_samples_per_second": 8.991,
+      "eval_steps_per_second": 1.135,
+      "step": 1026
+    },
+    {
+      "epoch": 1.003409644422796,
+      "grad_norm": 0.2860608398914337,
+      "learning_rate": 0.0002,
+      "loss": 1.7013,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0131514856307842,
+      "grad_norm": 0.3156210780143738,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0228933268387725,
+      "grad_norm": 0.26126575469970703,
+      "learning_rate": 0.0002,
+      "loss": 1.6182,
+      "step": 1050
+    },
+    {
+      "epoch": 1.032635168046761,
+      "grad_norm": 0.3019633889198303,
+      "learning_rate": 0.0002,
+      "loss": 1.7546,
+      "step": 1060
+    },
+    {
+      "epoch": 1.042377009254749,
+      "grad_norm": 0.534140944480896,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0521188504627375,
+      "grad_norm": 0.311872661113739,
+      "learning_rate": 0.0002,
+      "loss": 1.664,
+      "step": 1080
+    },
+    {
+      "epoch": 1.0618606916707258,
+      "grad_norm": 0.3276001513004303,
+      "learning_rate": 0.0002,
+      "loss": 1.7056,
+      "step": 1090
+    },
+    {
+      "epoch": 1.071602532878714,
+      "grad_norm": 0.35227468609809875,
+      "learning_rate": 0.0002,
+      "loss": 1.7943,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0813443740867024,
+      "grad_norm": 0.3597564995288849,
+      "learning_rate": 0.0002,
+      "loss": 1.7079,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0910862152946907,
+      "grad_norm": 0.3547225296497345,
+      "learning_rate": 0.0002,
+      "loss": 1.7635,
+      "step": 1120
+    },
+    {
+      "epoch": 1.100828056502679,
+      "grad_norm": 0.3399786353111267,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1105698977106673,
+      "grad_norm": 0.3309086263179779,
+      "learning_rate": 0.0002,
+      "loss": 1.7478,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1203117389186557,
+      "grad_norm": 0.39330706000328064,
+      "learning_rate": 0.0002,
+      "loss": 1.7372,
+      "step": 1150
+    },
+    {
+      "epoch": 1.130053580126644,
+      "grad_norm": 0.3628021776676178,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1397954213346322,
+      "grad_norm": 0.32995012402534485,
+      "learning_rate": 0.0002,
+      "loss": 1.8022,
+      "step": 1170
+    },
+    {
+      "epoch": 1.1495372625426206,
+      "grad_norm": 0.36292821168899536,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 1180
+    },
+    {
+      "epoch": 1.159279103750609,
+      "grad_norm": 0.3470092713832855,
+      "learning_rate": 0.0002,
+      "loss": 1.7156,
+      "step": 1190
+    },
+    {
+      "epoch": 1.169020944958597,
+      "grad_norm": 0.3496156334877014,
+      "learning_rate": 0.0002,
+      "loss": 1.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1787627861665855,
+      "grad_norm": 0.3442084789276123,
+      "learning_rate": 0.0002,
+      "loss": 1.716,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1885046273745739,
+      "grad_norm": 0.34983909130096436,
+      "learning_rate": 0.0002,
+      "loss": 1.7763,
+      "step": 1220
+    },
+    {
+      "epoch": 1.198246468582562,
+      "grad_norm": 0.36505937576293945,
+      "learning_rate": 0.0002,
+      "loss": 1.6964,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2079883097905504,
+      "grad_norm": 0.31624770164489746,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2177301509985388,
+      "grad_norm": 0.3528020679950714,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2274719922065271,
+      "grad_norm": 0.29294025897979736,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2372138334145153,
+      "grad_norm": 0.35048434138298035,
+      "learning_rate": 0.0002,
+      "loss": 1.719,
+      "step": 1270
+    },
+    {
+      "epoch": 1.2469556746225037,
+      "grad_norm": 0.35224461555480957,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 1280
+    },
+    {
+      "epoch": 1.256697515830492,
+      "grad_norm": 0.4041554629802704,
+      "learning_rate": 0.0002,
+      "loss": 1.6836,
+      "step": 1290
+    },
+    {
+      "epoch": 1.2664393570384802,
+      "grad_norm": 0.3447791039943695,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2761811982464686,
+      "grad_norm": 0.3315333425998688,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1310
+    },
+    {
+      "epoch": 1.285923039454457,
+      "grad_norm": 0.3587741255760193,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2956648806624451,
+      "grad_norm": 0.3704394996166229,
+      "learning_rate": 0.0002,
+      "loss": 1.6707,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3054067218704335,
+      "grad_norm": 0.38131803274154663,
+      "learning_rate": 0.0002,
+      "loss": 1.7374,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3151485630784219,
+      "grad_norm": 0.36109617352485657,
+      "learning_rate": 0.0002,
+      "loss": 1.5956,
+      "step": 1350
+    },
+    {
+      "epoch": 1.32489040428641,
+      "grad_norm": 0.37283554673194885,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3346322454943984,
+      "grad_norm": 0.31808891892433167,
+      "learning_rate": 0.0002,
+      "loss": 1.7351,
+      "step": 1370
+    },
+    {
+      "epoch": 1.3443740867023868,
+      "grad_norm": 0.3370385766029358,
+      "learning_rate": 0.0002,
+      "loss": 1.8368,
+      "step": 1380
+    },
+    {
+      "epoch": 1.354115927910375,
+      "grad_norm": 0.3568558394908905,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3638577691183633,
+      "grad_norm": 0.3537410497665405,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.3735996103263517,
+      "grad_norm": 0.3536544144153595,
+      "learning_rate": 0.0002,
+      "loss": 1.6534,
+      "step": 1410
+    },
+    {
+      "epoch": 1.38334145153434,
+      "grad_norm": 0.3772895038127899,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3930832927423282,
+      "grad_norm": 0.38079720735549927,
+      "learning_rate": 0.0002,
+      "loss": 1.7344,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4028251339503166,
+      "grad_norm": 0.3811109662055969,
+      "learning_rate": 0.0002,
+      "loss": 1.7121,
+      "step": 1440
+    },
+    {
+      "epoch": 1.412566975158305,
+      "grad_norm": 0.38586318492889404,
+      "learning_rate": 0.0002,
+      "loss": 1.6424,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4223088163662934,
+      "grad_norm": 0.3405744135379791,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 1460
+    },
+    {
+      "epoch": 1.4320506575742815,
+      "grad_norm": 0.39527642726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1470
+    },
+    {
+      "epoch": 1.44179249878227,
+      "grad_norm": 0.4494728744029999,
+      "learning_rate": 0.0002,
+      "loss": 1.7927,
+      "step": 1480
+    },
+    {
+      "epoch": 1.4515343399902583,
+      "grad_norm": 0.34068453311920166,
+      "learning_rate": 0.0002,
+      "loss": 1.5993,
+      "step": 1490
+    },
+    {
+      "epoch": 1.4612761811982464,
+      "grad_norm": 0.36169710755348206,
+      "learning_rate": 0.0002,
+      "loss": 1.73,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4710180224062348,
+      "grad_norm": 0.31519418954849243,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4807598636142232,
+      "grad_norm": 0.35117292404174805,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4905017048222113,
+      "grad_norm": 0.40951141715049744,
+      "learning_rate": 0.0002,
+      "loss": 1.6662,
+      "step": 1530
+    },
+    {
+      "epoch": 1.5002435460301997,
+      "grad_norm": 0.37542906403541565,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 1540
+    },
+    {
+      "epoch": 1.509985387238188,
+      "grad_norm": 0.35395753383636475,
+      "learning_rate": 0.0002,
+      "loss": 1.6563,
+      "step": 1550
+    },
+    {
+      "epoch": 1.5197272284461762,
+      "grad_norm": 0.35497018694877625,
+      "learning_rate": 0.0002,
+      "loss": 1.6517,
+      "step": 1560
+    },
+    {
+      "epoch": 1.5294690696541646,
+      "grad_norm": 0.3693031072616577,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 1570
+    },
+    {
+      "epoch": 1.539210910862153,
+      "grad_norm": 0.34013301134109497,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1580
+    },
+    {
+      "epoch": 1.5489527520701412,
+      "grad_norm": 0.37312784790992737,
+      "learning_rate": 0.0002,
+      "loss": 1.7364,
+      "step": 1590
+    },
+    {
+      "epoch": 1.5586945932781295,
+      "grad_norm": 0.357496440410614,
+      "learning_rate": 0.0002,
+      "loss": 1.6698,
+      "step": 1600
+    },
+    {
+      "epoch": 1.568436434486118,
+      "grad_norm": 0.35192370414733887,
+      "learning_rate": 0.0002,
+      "loss": 1.7113,
+      "step": 1610
+    },
+    {
+      "epoch": 1.578178275694106,
+      "grad_norm": 0.34144821763038635,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5879201169020944,
+      "grad_norm": 0.3320509195327759,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5976619581100828,
+      "grad_norm": 0.34178847074508667,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1640
+    },
+    {
+      "epoch": 1.607403799318071,
+      "grad_norm": 0.36567580699920654,
+      "learning_rate": 0.0002,
+      "loss": 1.7318,
+      "step": 1650
+    },
+    {
+      "epoch": 1.6171456405260596,
+      "grad_norm": 0.35599812865257263,
+      "learning_rate": 0.0002,
+      "loss": 1.6971,
+      "step": 1660
+    },
+    {
+      "epoch": 1.6268874817340477,
+      "grad_norm": 0.33765384554862976,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1670
+    },
+    {
+      "epoch": 1.636629322942036,
+      "grad_norm": 0.33142679929733276,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1680
+    },
+    {
+      "epoch": 1.6463711641500245,
+      "grad_norm": 0.6959079504013062,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1690
+    },
+    {
+      "epoch": 1.6561130053580126,
+      "grad_norm": 0.35073819756507874,
+      "learning_rate": 0.0002,
+      "loss": 1.7665,
+      "step": 1700
+    },
+    {
+      "epoch": 1.665854846566001,
+      "grad_norm": 0.3461478352546692,
+      "learning_rate": 0.0002,
+      "loss": 1.7571,
+      "step": 1710
+    },
+    {
+      "epoch": 1.6755966877739894,
+      "grad_norm": 0.3697752058506012,
+      "learning_rate": 0.0002,
+      "loss": 1.6608,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6853385289819776,
+      "grad_norm": 0.3755154609680176,
+      "learning_rate": 0.0002,
+      "loss": 1.729,
+      "step": 1730
+    },
+    {
+      "epoch": 1.695080370189966,
+      "grad_norm": 0.33977627754211426,
+      "learning_rate": 0.0002,
+      "loss": 1.6618,
+      "step": 1740
+    },
+    {
+      "epoch": 1.7048222113979543,
+      "grad_norm": 0.4001041650772095,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1750
+    },
+    {
+      "epoch": 1.7145640526059425,
+      "grad_norm": 0.36998286843299866,
+      "learning_rate": 0.0002,
+      "loss": 1.6756,
+      "step": 1760
+    },
+    {
+      "epoch": 1.7243058938139308,
+      "grad_norm": 0.39944565296173096,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1770
+    },
+    {
+      "epoch": 1.7340477350219192,
+      "grad_norm": 0.4002859890460968,
+      "learning_rate": 0.0002,
+      "loss": 1.6725,
+      "step": 1780
+    },
+    {
+      "epoch": 1.7437895762299074,
+      "grad_norm": 0.33336859941482544,
+      "learning_rate": 0.0002,
+      "loss": 1.768,
+      "step": 1790
+    },
+    {
+      "epoch": 1.7535314174378958,
+      "grad_norm": 0.35853952169418335,
+      "learning_rate": 0.0002,
+      "loss": 1.7816,
+      "step": 1800
+    },
+    {
+      "epoch": 1.7632732586458841,
+      "grad_norm": 0.35876700282096863,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 1810
+    },
+    {
+      "epoch": 1.7730150998538723,
+      "grad_norm": 0.3497968912124634,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7827569410618607,
+      "grad_norm": 0.33182016015052795,
+      "learning_rate": 0.0002,
+      "loss": 1.7128,
+      "step": 1830
+    },
+    {
+      "epoch": 1.792498782269849,
+      "grad_norm": 0.33359771966934204,
+      "learning_rate": 0.0002,
+      "loss": 1.7594,
+      "step": 1840
+    },
+    {
+      "epoch": 1.8022406234778372,
+      "grad_norm": 0.38070961833000183,
+      "learning_rate": 0.0002,
+      "loss": 1.8611,
+      "step": 1850
+    },
+    {
+      "epoch": 1.8119824646858256,
+      "grad_norm": 0.34111160039901733,
+      "learning_rate": 0.0002,
+      "loss": 1.7576,
+      "step": 1860
+    },
+    {
+      "epoch": 1.821724305893814,
+      "grad_norm": 0.4439302980899811,
+      "learning_rate": 0.0002,
+      "loss": 1.7533,
+      "step": 1870
+    },
+    {
+      "epoch": 1.8314661471018021,
+      "grad_norm": 0.37065210938453674,
+      "learning_rate": 0.0002,
+      "loss": 1.7931,
+      "step": 1880
+    },
+    {
+      "epoch": 1.8412079883097907,
+      "grad_norm": 0.33630406856536865,
+      "learning_rate": 0.0002,
+      "loss": 1.7392,
+      "step": 1890
+    },
+    {
+      "epoch": 1.8509498295177789,
+      "grad_norm": 0.334553986787796,
+      "learning_rate": 0.0002,
+      "loss": 1.6369,
+      "step": 1900
+    },
+    {
+      "epoch": 1.860691670725767,
+      "grad_norm": 0.3603808879852295,
+      "learning_rate": 0.0002,
+      "loss": 1.7271,
+      "step": 1910
+    },
+    {
+      "epoch": 1.8704335119337556,
+      "grad_norm": 0.4307343363761902,
+      "learning_rate": 0.0002,
+      "loss": 1.6777,
+      "step": 1920
+    },
+    {
+      "epoch": 1.8801753531417438,
+      "grad_norm": 0.455602765083313,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8899171943497322,
+      "grad_norm": 0.35242316126823425,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 1940
+    },
+    {
+      "epoch": 1.8996590355577205,
+      "grad_norm": 0.3589116632938385,
+      "learning_rate": 0.0002,
+      "loss": 1.7152,
+      "step": 1950
+    },
+    {
+      "epoch": 1.9094008767657087,
+      "grad_norm": 0.3540741801261902,
+      "learning_rate": 0.0002,
+      "loss": 1.7125,
+      "step": 1960
+    },
+    {
+      "epoch": 1.919142717973697,
+      "grad_norm": 0.3547612428665161,
+      "learning_rate": 0.0002,
+      "loss": 1.6873,
+      "step": 1970
+    },
+    {
+      "epoch": 1.9288845591816854,
+      "grad_norm": 0.3485773503780365,
+      "learning_rate": 0.0002,
+      "loss": 1.6995,
+      "step": 1980
+    },
+    {
+      "epoch": 1.9386264003896736,
+      "grad_norm": 0.3560304641723633,
+      "learning_rate": 0.0002,
+      "loss": 1.7301,
+      "step": 1990
+    },
+    {
+      "epoch": 1.948368241597662,
+      "grad_norm": 0.33299335837364197,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 2000
+    },
+    {
+      "epoch": 1.9581100828056504,
+      "grad_norm": 0.35622233152389526,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 2010
+    },
+    {
+      "epoch": 1.9678519240136385,
+      "grad_norm": 0.3681301474571228,
+      "learning_rate": 0.0002,
+      "loss": 1.7201,
+      "step": 2020
+    },
+    {
+      "epoch": 1.9775937652216269,
+      "grad_norm": 0.36158084869384766,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 2030
+    },
+    {
+      "epoch": 1.9873356064296153,
+      "grad_norm": 0.32560569047927856,
+      "learning_rate": 0.0002,
+      "loss": 1.6332,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9970774476376034,
+      "grad_norm": 0.37404149770736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6958,
+      "step": 2050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8119343519210815,
+      "eval_runtime": 96.0045,
+      "eval_samples_per_second": 5.281,
+      "eval_steps_per_second": 0.667,
+      "step": 2053
+    },
+    {
+      "epoch": 2.006819288845592,
+      "grad_norm": 0.374188631772995,
+      "learning_rate": 0.0002,
+      "loss": 1.657,
+      "step": 2060
+    },
+    {
+      "epoch": 2.01656113005358,
+      "grad_norm": 0.421764075756073,
+      "learning_rate": 0.0002,
+      "loss": 1.5655,
+      "step": 2070
+    },
+    {
+      "epoch": 2.0263029712615683,
+      "grad_norm": 0.43841829895973206,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2080
+    },
+    {
+      "epoch": 2.036044812469557,
+      "grad_norm": 0.42298218607902527,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 2090
+    },
+    {
+      "epoch": 2.045786653677545,
+      "grad_norm": 0.43669602274894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5883,
+      "step": 2100
+    },
+    {
+      "epoch": 2.0555284948855332,
+      "grad_norm": 0.4080469012260437,
+      "learning_rate": 0.0002,
+      "loss": 1.6143,
+      "step": 2110
+    },
+    {
+      "epoch": 2.065270336093522,
+      "grad_norm": 0.483192503452301,
+      "learning_rate": 0.0002,
+      "loss": 1.5591,
+      "step": 2120
+    },
+    {
+      "epoch": 2.07501217730151,
+      "grad_norm": 0.44427400827407837,
+      "learning_rate": 0.0002,
+      "loss": 1.6492,
+      "step": 2130
+    },
+    {
+      "epoch": 2.084754018509498,
+      "grad_norm": 0.48835131525993347,
+      "learning_rate": 0.0002,
+      "loss": 1.5845,
+      "step": 2140
+    },
+    {
+      "epoch": 2.0944958597174868,
+      "grad_norm": 0.42733684182167053,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 2150
+    },
+    {
+      "epoch": 2.104237700925475,
+      "grad_norm": 0.4258694648742676,
+      "learning_rate": 0.0002,
+      "loss": 1.5562,
+      "step": 2160
+    },
+    {
+      "epoch": 2.113979542133463,
+      "grad_norm": 0.5164985656738281,
+      "learning_rate": 0.0002,
+      "loss": 1.517,
+      "step": 2170
+    },
+    {
+      "epoch": 2.1237213833414517,
+      "grad_norm": 0.4279228150844574,
+      "learning_rate": 0.0002,
+      "loss": 1.6393,
+      "step": 2180
+    },
+    {
+      "epoch": 2.13346322454944,
+      "grad_norm": 0.48209506273269653,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2190
+    },
+    {
+      "epoch": 2.143205065757428,
+      "grad_norm": 0.4071785509586334,
+      "learning_rate": 0.0002,
+      "loss": 1.6142,
+      "step": 2200
+    },
+    {
+      "epoch": 2.1529469069654166,
+      "grad_norm": 0.4629398584365845,
+      "learning_rate": 0.0002,
+      "loss": 1.5469,
+      "step": 2210
+    },
+    {
+      "epoch": 2.1626887481734047,
+      "grad_norm": 0.44390997290611267,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2220
+    },
+    {
+      "epoch": 2.172430589381393,
+      "grad_norm": 0.46886971592903137,
+      "learning_rate": 0.0002,
+      "loss": 1.5395,
+      "step": 2230
+    },
+    {
+      "epoch": 2.1821724305893815,
+      "grad_norm": 0.43745434284210205,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 2240
+    },
+    {
+      "epoch": 2.1919142717973696,
+      "grad_norm": 0.42737245559692383,
+      "learning_rate": 0.0002,
+      "loss": 1.5416,
+      "step": 2250
+    },
+    {
+      "epoch": 2.201656113005358,
+      "grad_norm": 0.5028428435325623,
+      "learning_rate": 0.0002,
+      "loss": 1.627,
+      "step": 2260
+    },
+    {
+      "epoch": 2.2113979542133464,
+      "grad_norm": 0.48987212777137756,
+      "learning_rate": 0.0002,
+      "loss": 1.6148,
+      "step": 2270
+    },
+    {
+      "epoch": 2.2211397954213346,
+      "grad_norm": 0.48186370730400085,
+      "learning_rate": 0.0002,
+      "loss": 1.6107,
+      "step": 2280
+    },
+    {
+      "epoch": 2.2308816366293227,
+      "grad_norm": 0.4417429566383362,
+      "learning_rate": 0.0002,
+      "loss": 1.6657,
+      "step": 2290
+    },
+    {
+      "epoch": 2.2406234778373113,
+      "grad_norm": 0.4757710099220276,
+      "learning_rate": 0.0002,
+      "loss": 1.595,
+      "step": 2300
+    },
+    {
+      "epoch": 2.2503653190452995,
+      "grad_norm": 0.44449448585510254,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2310
+    },
+    {
+      "epoch": 2.260107160253288,
+      "grad_norm": 0.5070863962173462,
+      "learning_rate": 0.0002,
+      "loss": 1.5742,
+      "step": 2320
+    },
+    {
+      "epoch": 2.269849001461276,
+      "grad_norm": 0.4967133700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.5831,
+      "step": 2330
+    },
+    {
+      "epoch": 2.2795908426692644,
+      "grad_norm": 0.5110220909118652,
+      "learning_rate": 0.0002,
+      "loss": 1.5857,
+      "step": 2340
+    },
+    {
+      "epoch": 2.289332683877253,
+      "grad_norm": 0.47984135150909424,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2350
+    },
+    {
+      "epoch": 2.299074525085241,
+      "grad_norm": 0.5005794763565063,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 2360
+    },
+    {
+      "epoch": 2.3088163662932293,
+      "grad_norm": 0.4991425573825836,
+      "learning_rate": 0.0002,
+      "loss": 1.6131,
+      "step": 2370
+    },
+    {
+      "epoch": 2.318558207501218,
+      "grad_norm": 0.4948616623878479,
+      "learning_rate": 0.0002,
+      "loss": 1.5386,
+      "step": 2380
+    },
+    {
+      "epoch": 2.328300048709206,
+      "grad_norm": 0.4533160328865051,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 2390
+    },
+    {
+      "epoch": 2.338041889917194,
+      "grad_norm": 0.5871071219444275,
+      "learning_rate": 0.0002,
+      "loss": 1.5856,
+      "step": 2400
+    },
+    {
+      "epoch": 2.347783731125183,
+      "grad_norm": 0.5048075914382935,
+      "learning_rate": 0.0002,
+      "loss": 1.591,
+      "step": 2410
+    },
+    {
+      "epoch": 2.357525572333171,
+      "grad_norm": 0.4973750412464142,
+      "learning_rate": 0.0002,
+      "loss": 1.6165,
+      "step": 2420
+    },
+    {
+      "epoch": 2.367267413541159,
+      "grad_norm": 0.48294538259506226,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 2430
+    },
+    {
+      "epoch": 2.3770092547491477,
+      "grad_norm": 0.7180454134941101,
+      "learning_rate": 0.0002,
+      "loss": 1.5782,
+      "step": 2440
+    },
+    {
+      "epoch": 2.386751095957136,
+      "grad_norm": 0.4627632796764374,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 2450
+    },
+    {
+      "epoch": 2.396492937165124,
+      "grad_norm": 0.4834378957748413,
+      "learning_rate": 0.0002,
+      "loss": 1.586,
+      "step": 2460
+    },
+    {
+      "epoch": 2.4062347783731126,
+      "grad_norm": 0.5173670649528503,
+      "learning_rate": 0.0002,
+      "loss": 1.6145,
+      "step": 2470
+    },
+    {
+      "epoch": 2.4159766195811008,
+      "grad_norm": 0.49652737379074097,
+      "learning_rate": 0.0002,
+      "loss": 1.5464,
+      "step": 2480
+    },
+    {
+      "epoch": 2.4257184607890894,
+      "grad_norm": 0.47052669525146484,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 2490
+    },
+    {
+      "epoch": 2.4354603019970775,
+      "grad_norm": 0.5188006162643433,
+      "learning_rate": 0.0002,
+      "loss": 1.6215,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4452021432050657,
+      "grad_norm": 0.5010119676589966,
+      "learning_rate": 0.0002,
+      "loss": 1.5512,
+      "step": 2510
+    },
+    {
+      "epoch": 2.4549439844130543,
+      "grad_norm": 0.4765235483646393,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2520
+    },
+    {
+      "epoch": 2.4646858256210424,
+      "grad_norm": 0.5292699337005615,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2530
+    },
+    {
+      "epoch": 2.4744276668290306,
+      "grad_norm": 0.48555099964141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5866,
+      "step": 2540
+    },
+    {
+      "epoch": 2.484169508037019,
+      "grad_norm": 0.4764043092727661,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 2550
+    },
+    {
+      "epoch": 2.4939113492450073,
+      "grad_norm": 0.47839659452438354,
+      "learning_rate": 0.0002,
+      "loss": 1.6545,
+      "step": 2560
+    },
+    {
+      "epoch": 2.5036531904529955,
+      "grad_norm": 0.4514436721801758,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 2570
+    },
+    {
+      "epoch": 2.513395031660984,
+      "grad_norm": 0.5681955218315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6389,
+      "step": 2580
+    },
+    {
+      "epoch": 2.5231368728689723,
+      "grad_norm": 0.49655985832214355,
+      "learning_rate": 0.0002,
+      "loss": 1.5976,
+      "step": 2590
+    },
+    {
+      "epoch": 2.5328787140769604,
+      "grad_norm": 0.5077657103538513,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 2600
+    },
+    {
+      "epoch": 2.542620555284949,
+      "grad_norm": 0.5643279552459717,
+      "learning_rate": 0.0002,
+      "loss": 1.5658,
+      "step": 2610
+    },
+    {
+      "epoch": 2.552362396492937,
+      "grad_norm": 0.4715031087398529,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 2620
+    },
+    {
+      "epoch": 2.5621042377009253,
+      "grad_norm": 0.528400719165802,
+      "learning_rate": 0.0002,
+      "loss": 1.5907,
+      "step": 2630
+    },
+    {
+      "epoch": 2.571846078908914,
+      "grad_norm": 0.49469611048698425,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2640
+    },
+    {
+      "epoch": 2.581587920116902,
+      "grad_norm": 0.4567806124687195,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 2650
+    },
+    {
+      "epoch": 2.5913297613248902,
+      "grad_norm": 0.5357107520103455,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 2660
+    },
+    {
+      "epoch": 2.601071602532879,
+      "grad_norm": 0.46977677941322327,
+      "learning_rate": 0.0002,
+      "loss": 1.6078,
+      "step": 2670
+    },
+    {
+      "epoch": 2.610813443740867,
+      "grad_norm": 0.6626771092414856,
+      "learning_rate": 0.0002,
+      "loss": 1.6994,
+      "step": 2680
+    },
+    {
+      "epoch": 2.620555284948855,
+      "grad_norm": 0.4587472081184387,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2690
+    },
+    {
+      "epoch": 2.6302971261568437,
+      "grad_norm": 0.4816797077655792,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 2700
+    },
+    {
+      "epoch": 2.640038967364832,
+      "grad_norm": 0.4856809675693512,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 2710
+    },
+    {
+      "epoch": 2.64978080857282,
+      "grad_norm": 0.46010780334472656,
+      "learning_rate": 0.0002,
+      "loss": 1.563,
+      "step": 2720
+    },
+    {
+      "epoch": 2.6595226497808087,
+      "grad_norm": 0.4637954533100128,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 2730
+    },
+    {
+      "epoch": 2.669264490988797,
+      "grad_norm": 0.5954997539520264,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 2740
+    },
+    {
+      "epoch": 2.679006332196785,
+      "grad_norm": 0.5071861743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.5795,
+      "step": 2750
+    },
+    {
+      "epoch": 2.6887481734047736,
+      "grad_norm": 0.5415477156639099,
+      "learning_rate": 0.0002,
+      "loss": 1.573,
+      "step": 2760
+    },
+    {
+      "epoch": 2.6984900146127617,
+      "grad_norm": 0.5618549585342407,
+      "learning_rate": 0.0002,
+      "loss": 1.5476,
+      "step": 2770
+    },
+    {
+      "epoch": 2.70823185582075,
+      "grad_norm": 0.49338817596435547,
+      "learning_rate": 0.0002,
+      "loss": 1.608,
+      "step": 2780
+    },
+    {
+      "epoch": 2.7179736970287385,
+      "grad_norm": 0.5149586796760559,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2790
+    },
+    {
+      "epoch": 2.7277155382367266,
+      "grad_norm": 0.6247242093086243,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 2800
+    },
+    {
+      "epoch": 2.737457379444715,
+      "grad_norm": 0.4749542474746704,
+      "learning_rate": 0.0002,
+      "loss": 1.4655,
+      "step": 2810
+    },
+    {
+      "epoch": 2.7471992206527034,
+      "grad_norm": 0.4979191720485687,
+      "learning_rate": 0.0002,
+      "loss": 1.5984,
+      "step": 2820
+    },
+    {
+      "epoch": 2.7569410618606915,
+      "grad_norm": 0.4885074198246002,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2830
+    },
+    {
+      "epoch": 2.76668290306868,
+      "grad_norm": 0.5047747492790222,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 2840
+    },
+    {
+      "epoch": 2.7764247442766683,
+      "grad_norm": 0.5280140042304993,
+      "learning_rate": 0.0002,
+      "loss": 1.6574,
+      "step": 2850
+    },
+    {
+      "epoch": 2.7861665854846565,
+      "grad_norm": 0.477668434381485,
+      "learning_rate": 0.0002,
+      "loss": 1.5639,
+      "step": 2860
+    },
+    {
+      "epoch": 2.795908426692645,
+      "grad_norm": 0.4816327393054962,
+      "learning_rate": 0.0002,
+      "loss": 1.5923,
+      "step": 2870
+    },
+    {
+      "epoch": 2.805650267900633,
+      "grad_norm": 0.523259162902832,
+      "learning_rate": 0.0002,
+      "loss": 1.6377,
+      "step": 2880
+    },
+    {
+      "epoch": 2.8153921091086214,
+      "grad_norm": 0.5045270919799805,
+      "learning_rate": 0.0002,
+      "loss": 1.5779,
+      "step": 2890
+    },
+    {
+      "epoch": 2.82513395031661,
+      "grad_norm": 0.47986042499542236,
+      "learning_rate": 0.0002,
+      "loss": 1.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 2.834875791524598,
+      "grad_norm": 0.4858797490596771,
+      "learning_rate": 0.0002,
+      "loss": 1.5997,
+      "step": 2910
+    },
+    {
+      "epoch": 2.8446176327325867,
+      "grad_norm": 0.5261512398719788,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 2920
+    },
+    {
+      "epoch": 2.854359473940575,
+      "grad_norm": 0.630550742149353,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 2930
+    },
+    {
+      "epoch": 2.864101315148563,
+      "grad_norm": 0.49119752645492554,
+      "learning_rate": 0.0002,
+      "loss": 1.5889,
+      "step": 2940
+    },
+    {
+      "epoch": 2.8738431563565516,
+      "grad_norm": 0.4779070317745209,
+      "learning_rate": 0.0002,
+      "loss": 1.6033,
+      "step": 2950
+    },
+    {
+      "epoch": 2.88358499756454,
+      "grad_norm": 0.5059782266616821,
+      "learning_rate": 0.0002,
+      "loss": 1.6353,
+      "step": 2960
+    },
+    {
+      "epoch": 2.893326838772528,
+      "grad_norm": 0.5466655492782593,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2970
+    },
+    {
+      "epoch": 2.9030686799805165,
+      "grad_norm": 0.4865640103816986,
+      "learning_rate": 0.0002,
+      "loss": 1.619,
+      "step": 2980
+    },
+    {
+      "epoch": 2.9128105211885047,
+      "grad_norm": 0.6101235151290894,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 2990
+    },
+    {
+      "epoch": 2.922552362396493,
+      "grad_norm": 0.9112305641174316,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9322942036044815,
+      "grad_norm": 0.3938814103603363,
+      "learning_rate": 0.0002,
+      "loss": 1.6493,
+      "step": 3010
+    },
+    {
+      "epoch": 2.9420360448124696,
+      "grad_norm": 0.5500800609588623,
+      "learning_rate": 0.0002,
+      "loss": 1.5446,
+      "step": 3020
+    },
+    {
+      "epoch": 2.9517778860204578,
+      "grad_norm": 0.5346390604972839,
+      "learning_rate": 0.0002,
+      "loss": 1.5626,
+      "step": 3030
+    },
+    {
+      "epoch": 2.9615197272284464,
+      "grad_norm": 0.5245014429092407,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3040
+    },
+    {
+      "epoch": 2.9712615684364345,
+      "grad_norm": 0.4906884431838989,
+      "learning_rate": 0.0002,
+      "loss": 1.5549,
+      "step": 3050
+    },
+    {
+      "epoch": 2.9810034096444227,
+      "grad_norm": 0.47086769342422485,
+      "learning_rate": 0.0002,
+      "loss": 1.5796,
+      "step": 3060
+    },
+    {
+      "epoch": 2.9907452508524113,
+      "grad_norm": 0.5290229320526123,
+      "learning_rate": 0.0002,
+      "loss": 1.5917,
+      "step": 3070
+    },
+    {
+      "epoch": 2.9995129079396006,
+      "eval_loss": 1.8463934659957886,
+      "eval_runtime": 56.2401,
+      "eval_samples_per_second": 9.015,
+      "eval_steps_per_second": 1.138,
+      "step": 3079
+    },
+    {
+      "epoch": 3.0004870920603994,
+      "grad_norm": 0.49992576241493225,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 3080
+    },
+    {
+      "epoch": 3.0102289332683876,
+      "grad_norm": 0.8242783546447754,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3090
+    },
+    {
+      "epoch": 3.019970774476376,
+      "grad_norm": 0.6330569386482239,
+      "learning_rate": 0.0002,
+      "loss": 1.394,
+      "step": 3100
+    },
+    {
+      "epoch": 3.0297126156843643,
+      "grad_norm": 0.566097617149353,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 3110
+    },
+    {
+      "epoch": 3.0394544568923525,
+      "grad_norm": 0.6337586045265198,
+      "learning_rate": 0.0002,
+      "loss": 1.4365,
+      "step": 3120
+    },
+    {
+      "epoch": 3.049196298100341,
+      "grad_norm": 0.7339403033256531,
+      "learning_rate": 0.0002,
+      "loss": 1.3916,
+      "step": 3130
+    },
+    {
+      "epoch": 3.0589381393083293,
+      "grad_norm": 0.7187346816062927,
+      "learning_rate": 0.0002,
+      "loss": 1.4617,
+      "step": 3140
+    },
+    {
+      "epoch": 3.0686799805163174,
+      "grad_norm": 0.7116255760192871,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 3150
+    },
+    {
+      "epoch": 3.078421821724306,
+      "grad_norm": 0.6493807435035706,
+      "learning_rate": 0.0002,
+      "loss": 1.4452,
+      "step": 3160
+    },
+    {
+      "epoch": 3.088163662932294,
+      "grad_norm": 0.6777266263961792,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 3170
+    },
+    {
+      "epoch": 3.0979055041402823,
+      "grad_norm": 0.6342006325721741,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 3180
+    },
+    {
+      "epoch": 3.107647345348271,
+      "grad_norm": 0.6608964204788208,
+      "learning_rate": 0.0002,
+      "loss": 1.4748,
+      "step": 3190
+    },
+    {
+      "epoch": 3.117389186556259,
+      "grad_norm": 0.7230247259140015,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 3200
+    },
+    {
+      "epoch": 3.1271310277642472,
+      "grad_norm": 0.650368332862854,
+      "learning_rate": 0.0002,
+      "loss": 1.4049,
+      "step": 3210
+    },
+    {
+      "epoch": 3.136872868972236,
+      "grad_norm": 0.7319342494010925,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 3220
+    },
+    {
+      "epoch": 3.146614710180224,
+      "grad_norm": 0.7159963846206665,
+      "learning_rate": 0.0002,
+      "loss": 1.3872,
+      "step": 3230
+    },
+    {
+      "epoch": 3.156356551388212,
+      "grad_norm": 0.8905230164527893,
+      "learning_rate": 0.0002,
+      "loss": 1.5076,
+      "step": 3240
+    },
+    {
+      "epoch": 3.1660983925962007,
+      "grad_norm": 0.6920804381370544,
+      "learning_rate": 0.0002,
+      "loss": 1.3161,
+      "step": 3250
+    },
+    {
+      "epoch": 3.175840233804189,
+      "grad_norm": 0.6782063841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.3786,
+      "step": 3260
+    },
+    {
+      "epoch": 3.1855820750121775,
+      "grad_norm": 0.735325276851654,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 3270
+    },
+    {
+      "epoch": 3.1953239162201656,
+      "grad_norm": 0.6657978296279907,
+      "learning_rate": 0.0002,
+      "loss": 1.4027,
+      "step": 3280
+    },
+    {
+      "epoch": 3.205065757428154,
+      "grad_norm": 0.771315336227417,
+      "learning_rate": 0.0002,
+      "loss": 1.3456,
+      "step": 3290
+    },
+    {
+      "epoch": 3.2148075986361424,
+      "grad_norm": 0.6492983102798462,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 3300
+    },
+    {
+      "epoch": 3.2245494398441306,
+      "grad_norm": 0.7513770461082458,
+      "learning_rate": 0.0002,
+      "loss": 1.4125,
+      "step": 3310
+    },
+    {
+      "epoch": 3.2342912810521187,
+      "grad_norm": 0.7091423869132996,
+      "learning_rate": 0.0002,
+      "loss": 1.4032,
+      "step": 3320
+    },
+    {
+      "epoch": 3.2440331222601073,
+      "grad_norm": 0.6663975119590759,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 3330
+    },
+    {
+      "epoch": 3.2537749634680955,
+      "grad_norm": 0.6813122034072876,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 3340
+    },
+    {
+      "epoch": 3.2635168046760836,
+      "grad_norm": 0.6602569818496704,
+      "learning_rate": 0.0002,
+      "loss": 1.3681,
+      "step": 3350
+    },
+    {
+      "epoch": 3.2732586458840722,
+      "grad_norm": 0.718270480632782,
+      "learning_rate": 0.0002,
+      "loss": 1.4533,
+      "step": 3360
+    },
+    {
+      "epoch": 3.2830004870920604,
+      "grad_norm": 0.6884173154830933,
+      "learning_rate": 0.0002,
+      "loss": 1.4076,
+      "step": 3370
+    },
+    {
+      "epoch": 3.2927423283000485,
+      "grad_norm": 0.7039775848388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3380
+    },
+    {
+      "epoch": 3.302484169508037,
+      "grad_norm": 0.7444299459457397,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3390
+    },
+    {
+      "epoch": 3.3122260107160253,
+      "grad_norm": 0.7187064290046692,
+      "learning_rate": 0.0002,
+      "loss": 1.4255,
+      "step": 3400
+    },
+    {
+      "epoch": 3.3219678519240134,
+      "grad_norm": 0.599396288394928,
+      "learning_rate": 0.0002,
+      "loss": 1.3684,
+      "step": 3410
+    },
+    {
+      "epoch": 3.331709693132002,
+      "grad_norm": 0.7670390009880066,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3420
+    },
+    {
+      "epoch": 3.34145153433999,
+      "grad_norm": 0.6654478311538696,
+      "learning_rate": 0.0002,
+      "loss": 1.4411,
+      "step": 3430
+    },
+    {
+      "epoch": 3.351193375547979,
+      "grad_norm": 0.6644385457038879,
+      "learning_rate": 0.0002,
+      "loss": 1.4257,
+      "step": 3440
+    },
+    {
+      "epoch": 3.360935216755967,
+      "grad_norm": 0.6974098086357117,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3450
+    },
+    {
+      "epoch": 3.370677057963955,
+      "grad_norm": 0.7350399494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 3460
+    },
+    {
+      "epoch": 3.3804188991719437,
+      "grad_norm": 0.714721143245697,
+      "learning_rate": 0.0002,
+      "loss": 1.4176,
+      "step": 3470
+    },
+    {
+      "epoch": 3.390160740379932,
+      "grad_norm": 0.7006027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 3480
+    },
+    {
+      "epoch": 3.39990258158792,
+      "grad_norm": 0.6767925024032593,
+      "learning_rate": 0.0002,
+      "loss": 1.4888,
+      "step": 3490
+    },
+    {
+      "epoch": 3.4096444227959086,
+      "grad_norm": 0.6721355319023132,
+      "learning_rate": 0.0002,
+      "loss": 1.4116,
+      "step": 3500
+    },
+    {
+      "epoch": 3.419386264003897,
+      "grad_norm": 0.6845725178718567,
+      "learning_rate": 0.0002,
+      "loss": 1.443,
+      "step": 3510
+    },
+    {
+      "epoch": 3.429128105211885,
+      "grad_norm": 0.6882196664810181,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 3520
+    },
+    {
+      "epoch": 3.4388699464198735,
+      "grad_norm": 0.7663240432739258,
+      "learning_rate": 0.0002,
+      "loss": 1.4962,
+      "step": 3530
+    },
+    {
+      "epoch": 3.4486117876278617,
+      "grad_norm": 0.6304219365119934,
+      "learning_rate": 0.0002,
+      "loss": 1.4644,
+      "step": 3540
+    },
+    {
+      "epoch": 3.45835362883585,
+      "grad_norm": 0.668678879737854,
+      "learning_rate": 0.0002,
+      "loss": 1.4918,
+      "step": 3550
+    },
+    {
+      "epoch": 3.4680954700438384,
+      "grad_norm": 0.7526912093162537,
+      "learning_rate": 0.0002,
+      "loss": 1.4874,
+      "step": 3560
+    },
+    {
+      "epoch": 3.4778373112518266,
+      "grad_norm": 1.089495301246643,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 3570
+    },
+    {
+      "epoch": 3.4875791524598148,
+      "grad_norm": 0.7282902002334595,
+      "learning_rate": 0.0002,
+      "loss": 1.3871,
+      "step": 3580
+    },
+    {
+      "epoch": 3.4973209936678034,
+      "grad_norm": 0.6540156602859497,
+      "learning_rate": 0.0002,
+      "loss": 1.5077,
+      "step": 3590
+    },
+    {
+      "epoch": 3.5070628348757915,
+      "grad_norm": 0.6449568867683411,
+      "learning_rate": 0.0002,
+      "loss": 1.4367,
+      "step": 3600
+    },
+    {
+      "epoch": 3.5168046760837797,
+      "grad_norm": 0.7262216210365295,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 3610
+    },
+    {
+      "epoch": 3.5265465172917683,
+      "grad_norm": 0.6048615574836731,
+      "learning_rate": 0.0002,
+      "loss": 1.4374,
+      "step": 3620
+    },
+    {
+      "epoch": 3.5362883584997564,
+      "grad_norm": 0.6780537366867065,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 3630
+    },
+    {
+      "epoch": 3.5460301997077446,
+      "grad_norm": 0.6851925253868103,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 3640
+    },
+    {
+      "epoch": 3.555772040915733,
+      "grad_norm": 0.6530634164810181,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 3650
+    },
+    {
+      "epoch": 3.5655138821237213,
+      "grad_norm": 0.7193992733955383,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 3660
+    },
+    {
+      "epoch": 3.5752557233317095,
+      "grad_norm": 0.767496645450592,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 3670
+    },
+    {
+      "epoch": 3.584997564539698,
+      "grad_norm": 0.6912919282913208,
+      "learning_rate": 0.0002,
+      "loss": 1.4824,
+      "step": 3680
+    },
+    {
+      "epoch": 3.5947394057476862,
+      "grad_norm": 0.7383436560630798,
+      "learning_rate": 0.0002,
+      "loss": 1.4497,
+      "step": 3690
+    },
+    {
+      "epoch": 3.6044812469556744,
+      "grad_norm": 0.6746662855148315,
+      "learning_rate": 0.0002,
+      "loss": 1.4822,
+      "step": 3700
+    },
+    {
+      "epoch": 3.614223088163663,
+      "grad_norm": 0.6885138750076294,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 3710
+    },
+    {
+      "epoch": 3.623964929371651,
+      "grad_norm": 0.6694392561912537,
+      "learning_rate": 0.0002,
+      "loss": 1.4044,
+      "step": 3720
+    },
+    {
+      "epoch": 3.6337067705796393,
+      "grad_norm": 0.812358021736145,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 3730
+    },
+    {
+      "epoch": 3.643448611787628,
+      "grad_norm": 0.7267130017280579,
+      "learning_rate": 0.0002,
+      "loss": 1.4603,
+      "step": 3740
+    },
+    {
+      "epoch": 3.653190452995616,
+      "grad_norm": 0.6958749294281006,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 3750
+    },
+    {
+      "epoch": 3.6629322942036042,
+      "grad_norm": 0.6805673241615295,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3760
+    },
+    {
+      "epoch": 3.672674135411593,
+      "grad_norm": 0.7184410095214844,
+      "learning_rate": 0.0002,
+      "loss": 1.4338,
+      "step": 3770
+    },
+    {
+      "epoch": 3.682415976619581,
+      "grad_norm": 0.7716330289840698,
+      "learning_rate": 0.0002,
+      "loss": 1.3935,
+      "step": 3780
+    },
+    {
+      "epoch": 3.6921578178275696,
+      "grad_norm": 0.6675831079483032,
+      "learning_rate": 0.0002,
+      "loss": 1.384,
+      "step": 3790
+    },
+    {
+      "epoch": 3.7018996590355577,
+      "grad_norm": 0.6480095386505127,
+      "learning_rate": 0.0002,
+      "loss": 1.401,
+      "step": 3800
+    },
+    {
+      "epoch": 3.711641500243546,
+      "grad_norm": 0.6559418439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.5303,
+      "step": 3810
+    },
+    {
+      "epoch": 3.7213833414515345,
+      "grad_norm": 0.6596545577049255,
+      "learning_rate": 0.0002,
+      "loss": 1.4341,
+      "step": 3820
+    },
+    {
+      "epoch": 3.7311251826595226,
+      "grad_norm": 0.7172950506210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 3830
+    },
+    {
+      "epoch": 3.740867023867511,
+      "grad_norm": 0.796148419380188,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 3840
+    },
+    {
+      "epoch": 3.7506088650754994,
+      "grad_norm": 0.6600322723388672,
+      "learning_rate": 0.0002,
+      "loss": 1.4992,
+      "step": 3850
+    },
+    {
+      "epoch": 3.7603507062834876,
+      "grad_norm": 0.6776387691497803,
+      "learning_rate": 0.0002,
+      "loss": 1.4201,
+      "step": 3860
+    },
+    {
+      "epoch": 3.770092547491476,
+      "grad_norm": 0.7768304347991943,
+      "learning_rate": 0.0002,
+      "loss": 1.3893,
+      "step": 3870
+    },
+    {
+      "epoch": 3.7798343886994643,
+      "grad_norm": 1.0579794645309448,
+      "learning_rate": 0.0002,
+      "loss": 1.4886,
+      "step": 3880
+    },
+    {
+      "epoch": 3.7895762299074525,
+      "grad_norm": 0.6757252812385559,
+      "learning_rate": 0.0002,
+      "loss": 1.4556,
+      "step": 3890
+    },
+    {
+      "epoch": 3.799318071115441,
+      "grad_norm": 0.6706996560096741,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 3900
+    },
+    {
+      "epoch": 3.809059912323429,
+      "grad_norm": 0.7026948928833008,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 3910
+    },
+    {
+      "epoch": 3.8188017535314174,
+      "grad_norm": 0.6437768340110779,
+      "learning_rate": 0.0002,
+      "loss": 1.5487,
+      "step": 3920
+    },
+    {
+      "epoch": 3.828543594739406,
+      "grad_norm": 0.7015706300735474,
+      "learning_rate": 0.0002,
+      "loss": 1.4678,
+      "step": 3930
+    },
+    {
+      "epoch": 3.838285435947394,
+      "grad_norm": 0.7049482464790344,
+      "learning_rate": 0.0002,
+      "loss": 1.4891,
+      "step": 3940
+    },
+    {
+      "epoch": 3.8480272771553823,
+      "grad_norm": 0.6533724665641785,
+      "learning_rate": 0.0002,
+      "loss": 1.4208,
+      "step": 3950
+    },
+    {
+      "epoch": 3.857769118363371,
+      "grad_norm": 0.7312499284744263,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 3960
+    },
+    {
+      "epoch": 3.867510959571359,
+      "grad_norm": 0.6858801245689392,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3970
+    },
+    {
+      "epoch": 3.877252800779347,
+      "grad_norm": 0.770423173904419,
+      "learning_rate": 0.0002,
+      "loss": 1.4423,
+      "step": 3980
+    },
+    {
+      "epoch": 3.886994641987336,
+      "grad_norm": 0.6987539529800415,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 3990
+    },
+    {
+      "epoch": 3.896736483195324,
+      "grad_norm": 0.7072722315788269,
+      "learning_rate": 0.0002,
+      "loss": 1.4791,
+      "step": 4000
+    },
+    {
+      "epoch": 3.906478324403312,
+      "grad_norm": 0.6492931842803955,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4010
+    },
+    {
+      "epoch": 3.9162201656113007,
+      "grad_norm": 0.7716232538223267,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 4020
+    },
+    {
+      "epoch": 3.925962006819289,
+      "grad_norm": 0.722949743270874,
+      "learning_rate": 0.0002,
+      "loss": 1.4758,
+      "step": 4030
+    },
+    {
+      "epoch": 3.935703848027277,
+      "grad_norm": 0.7434365749359131,
+      "learning_rate": 0.0002,
+      "loss": 1.3914,
+      "step": 4040
+    },
+    {
+      "epoch": 3.9454456892352656,
+      "grad_norm": 0.6691509485244751,
+      "learning_rate": 0.0002,
+      "loss": 1.4763,
+      "step": 4050
+    },
+    {
+      "epoch": 3.9551875304432538,
+      "grad_norm": 0.6850284337997437,
+      "learning_rate": 0.0002,
+      "loss": 1.4555,
+      "step": 4060
+    },
+    {
+      "epoch": 3.964929371651242,
+      "grad_norm": 0.6954452991485596,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4070
+    },
+    {
+      "epoch": 3.9746712128592305,
+      "grad_norm": 0.9316364526748657,
+      "learning_rate": 0.0002,
+      "loss": 1.417,
+      "step": 4080
+    },
+    {
+      "epoch": 3.9844130540672187,
+      "grad_norm": 0.6908289194107056,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 4090
+    },
+    {
+      "epoch": 3.994154895275207,
+      "grad_norm": 0.666782021522522,
+      "learning_rate": 0.0002,
+      "loss": 1.4404,
+      "step": 4100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9233275651931763,
+      "eval_runtime": 55.9536,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 1.144,
+      "step": 4106
+    },
+    {
+      "epoch": 4.003896736483195,
+      "grad_norm": 0.7726166248321533,
+      "learning_rate": 0.0002,
+      "loss": 1.3489,
+      "step": 4110
+    },
+    {
+      "epoch": 4.013638577691184,
+      "grad_norm": 1.1338967084884644,
+      "learning_rate": 0.0002,
+      "loss": 1.1415,
+      "step": 4120
+    },
+    {
+      "epoch": 4.023380418899172,
+      "grad_norm": 0.9530029296875,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 4130
+    },
+    {
+      "epoch": 4.03312226010716,
+      "grad_norm": 1.1058554649353027,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 4140
+    },
+    {
+      "epoch": 4.042864101315149,
+      "grad_norm": 0.8765049576759338,
+      "learning_rate": 0.0002,
+      "loss": 1.2381,
+      "step": 4150
+    },
+    {
+      "epoch": 4.052605942523137,
+      "grad_norm": 1.1774667501449585,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 4160
+    },
+    {
+      "epoch": 4.062347783731125,
+      "grad_norm": 0.9301433563232422,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 4170
+    },
+    {
+      "epoch": 4.072089624939114,
+      "grad_norm": 1.0196778774261475,
+      "learning_rate": 0.0002,
+      "loss": 1.1807,
+      "step": 4180
+    },
+    {
+      "epoch": 4.081831466147102,
+      "grad_norm": 1.1380577087402344,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 4190
+    },
+    {
+      "epoch": 4.09157330735509,
+      "grad_norm": 0.9121319651603699,
+      "learning_rate": 0.0002,
+      "loss": 1.2521,
+      "step": 4200
+    },
+    {
+      "epoch": 4.101315148563079,
+      "grad_norm": 0.9495378732681274,
+      "learning_rate": 0.0002,
+      "loss": 1.1747,
+      "step": 4210
+    },
+    {
+      "epoch": 4.1110569897710665,
+      "grad_norm": 0.8058680295944214,
+      "learning_rate": 0.0002,
+      "loss": 1.1829,
+      "step": 4220
+    },
+    {
+      "epoch": 4.120798830979055,
+      "grad_norm": 1.000887393951416,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 4230
+    },
+    {
+      "epoch": 4.130540672187044,
+      "grad_norm": 0.9529102444648743,
+      "learning_rate": 0.0002,
+      "loss": 1.1947,
+      "step": 4240
+    },
+    {
+      "epoch": 4.140282513395031,
+      "grad_norm": 1.0257115364074707,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 4250
+    },
+    {
+      "epoch": 4.15002435460302,
+      "grad_norm": 0.9590303897857666,
+      "learning_rate": 0.0002,
+      "loss": 1.2293,
+      "step": 4260
+    },
+    {
+      "epoch": 4.159766195811009,
+      "grad_norm": 1.065291166305542,
+      "learning_rate": 0.0002,
+      "loss": 1.1918,
+      "step": 4270
+    },
+    {
+      "epoch": 4.169508037018996,
+      "grad_norm": 0.8819697499275208,
+      "learning_rate": 0.0002,
+      "loss": 1.2323,
+      "step": 4280
+    },
+    {
+      "epoch": 4.179249878226985,
+      "grad_norm": 1.0335261821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.2167,
+      "step": 4290
+    },
+    {
+      "epoch": 4.1889917194349735,
+      "grad_norm": 0.8872809410095215,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 4300
+    },
+    {
+      "epoch": 4.198733560642961,
+      "grad_norm": 0.9883159399032593,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 4310
+    },
+    {
+      "epoch": 4.20847540185095,
+      "grad_norm": 1.0254192352294922,
+      "learning_rate": 0.0002,
+      "loss": 1.2544,
+      "step": 4320
+    },
+    {
+      "epoch": 4.218217243058938,
+      "grad_norm": 0.9432600736618042,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 4330
+    },
+    {
+      "epoch": 4.227959084266926,
+      "grad_norm": 1.1008676290512085,
+      "learning_rate": 0.0002,
+      "loss": 1.2684,
+      "step": 4340
+    },
+    {
+      "epoch": 4.237700925474915,
+      "grad_norm": 1.0829699039459229,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 4350
+    },
+    {
+      "epoch": 4.247442766682903,
+      "grad_norm": 1.016847848892212,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4360
+    },
+    {
+      "epoch": 4.257184607890891,
+      "grad_norm": 0.8924864530563354,
+      "learning_rate": 0.0002,
+      "loss": 1.2375,
+      "step": 4370
+    },
+    {
+      "epoch": 4.26692644909888,
+      "grad_norm": 0.9300530552864075,
+      "learning_rate": 0.0002,
+      "loss": 1.1987,
+      "step": 4380
+    },
+    {
+      "epoch": 4.276668290306868,
+      "grad_norm": 0.9684814810752869,
+      "learning_rate": 0.0002,
+      "loss": 1.1696,
+      "step": 4390
+    },
+    {
+      "epoch": 4.286410131514856,
+      "grad_norm": 0.9916250705718994,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 4400
+    },
+    {
+      "epoch": 4.2961519727228445,
+      "grad_norm": 0.903680145740509,
+      "learning_rate": 0.0002,
+      "loss": 1.2402,
+      "step": 4410
+    },
+    {
+      "epoch": 4.305893813930833,
+      "grad_norm": 0.8713505268096924,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 4420
+    },
+    {
+      "epoch": 4.315635655138821,
+      "grad_norm": 0.9983905553817749,
+      "learning_rate": 0.0002,
+      "loss": 1.1957,
+      "step": 4430
+    },
+    {
+      "epoch": 4.3253774963468095,
+      "grad_norm": 1.1689040660858154,
+      "learning_rate": 0.0002,
+      "loss": 1.2676,
+      "step": 4440
+    },
+    {
+      "epoch": 4.335119337554798,
+      "grad_norm": 0.9316853880882263,
+      "learning_rate": 0.0002,
+      "loss": 1.2166,
+      "step": 4450
+    },
+    {
+      "epoch": 4.344861178762786,
+      "grad_norm": 0.9175887107849121,
+      "learning_rate": 0.0002,
+      "loss": 1.222,
+      "step": 4460
+    },
+    {
+      "epoch": 4.354603019970774,
+      "grad_norm": 0.9348906874656677,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 4470
+    },
+    {
+      "epoch": 4.364344861178763,
+      "grad_norm": 0.9727016687393188,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 4480
+    },
+    {
+      "epoch": 4.374086702386751,
+      "grad_norm": 0.9843429923057556,
+      "learning_rate": 0.0002,
+      "loss": 1.2616,
+      "step": 4490
+    },
+    {
+      "epoch": 4.383828543594739,
+      "grad_norm": 0.9615852236747742,
+      "learning_rate": 0.0002,
+      "loss": 1.2488,
+      "step": 4500
+    },
+    {
+      "epoch": 4.393570384802728,
+      "grad_norm": 0.9688583612442017,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 4510
+    },
+    {
+      "epoch": 4.403312226010716,
+      "grad_norm": 0.9933668375015259,
+      "learning_rate": 0.0002,
+      "loss": 1.2546,
+      "step": 4520
+    },
+    {
+      "epoch": 4.413054067218704,
+      "grad_norm": 1.0626686811447144,
+      "learning_rate": 0.0002,
+      "loss": 1.2355,
+      "step": 4530
+    },
+    {
+      "epoch": 4.422795908426693,
+      "grad_norm": 0.9536267518997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 4540
+    },
+    {
+      "epoch": 4.432537749634681,
+      "grad_norm": 0.9777140021324158,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4550
+    },
+    {
+      "epoch": 4.442279590842669,
+      "grad_norm": 0.980780839920044,
+      "learning_rate": 0.0002,
+      "loss": 1.2878,
+      "step": 4560
+    },
+    {
+      "epoch": 4.452021432050658,
+      "grad_norm": 1.0147196054458618,
+      "learning_rate": 0.0002,
+      "loss": 1.2597,
+      "step": 4570
+    },
+    {
+      "epoch": 4.461763273258645,
+      "grad_norm": 0.9763361811637878,
+      "learning_rate": 0.0002,
+      "loss": 1.2148,
+      "step": 4580
+    },
+    {
+      "epoch": 4.471505114466634,
+      "grad_norm": 1.0300798416137695,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 4590
+    },
+    {
+      "epoch": 4.481246955674623,
+      "grad_norm": 0.8833121657371521,
+      "learning_rate": 0.0002,
+      "loss": 1.2665,
+      "step": 4600
+    },
+    {
+      "epoch": 4.490988796882611,
+      "grad_norm": 1.1214020252227783,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 4610
+    },
+    {
+      "epoch": 4.500730638090599,
+      "grad_norm": 0.8843787908554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2579,
+      "step": 4620
+    },
+    {
+      "epoch": 4.5104724792985875,
+      "grad_norm": 0.9942020773887634,
+      "learning_rate": 0.0002,
+      "loss": 1.2633,
+      "step": 4630
+    },
+    {
+      "epoch": 4.520214320506576,
+      "grad_norm": 1.0033202171325684,
+      "learning_rate": 0.0002,
+      "loss": 1.3172,
+      "step": 4640
+    },
+    {
+      "epoch": 4.529956161714564,
+      "grad_norm": 0.8767235279083252,
+      "learning_rate": 0.0002,
+      "loss": 1.2024,
+      "step": 4650
+    },
+    {
+      "epoch": 4.539698002922552,
+      "grad_norm": 1.0117276906967163,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 4660
+    },
+    {
+      "epoch": 4.549439844130541,
+      "grad_norm": 1.2787362337112427,
+      "learning_rate": 0.0002,
+      "loss": 1.2911,
+      "step": 4670
+    },
+    {
+      "epoch": 4.559181685338529,
+      "grad_norm": 0.8824878931045532,
+      "learning_rate": 0.0002,
+      "loss": 1.2603,
+      "step": 4680
+    },
+    {
+      "epoch": 4.568923526546517,
+      "grad_norm": 0.9209560751914978,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 4690
+    },
+    {
+      "epoch": 4.578665367754506,
+      "grad_norm": 1.1064010858535767,
+      "learning_rate": 0.0002,
+      "loss": 1.1916,
+      "step": 4700
+    },
+    {
+      "epoch": 4.588407208962494,
+      "grad_norm": 0.8914572596549988,
+      "learning_rate": 0.0002,
+      "loss": 1.2217,
+      "step": 4710
+    },
+    {
+      "epoch": 4.598149050170482,
+      "grad_norm": 1.0412265062332153,
+      "learning_rate": 0.0002,
+      "loss": 1.2861,
+      "step": 4720
+    },
+    {
+      "epoch": 4.607890891378471,
+      "grad_norm": 1.1950221061706543,
+      "learning_rate": 0.0002,
+      "loss": 1.262,
+      "step": 4730
+    },
+    {
+      "epoch": 4.617632732586459,
+      "grad_norm": 0.8938062787055969,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 4740
+    },
+    {
+      "epoch": 4.627374573794447,
+      "grad_norm": 0.9849569201469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 4750
+    },
+    {
+      "epoch": 4.637116415002436,
+      "grad_norm": 1.0081515312194824,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 4760
+    },
+    {
+      "epoch": 4.6468582562104235,
+      "grad_norm": 0.8566309213638306,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 4770
+    },
+    {
+      "epoch": 4.656600097418412,
+      "grad_norm": 1.1750118732452393,
+      "learning_rate": 0.0002,
+      "loss": 1.2723,
+      "step": 4780
+    },
+    {
+      "epoch": 4.666341938626401,
+      "grad_norm": 0.925502598285675,
+      "learning_rate": 0.0002,
+      "loss": 1.2537,
+      "step": 4790
+    },
+    {
+      "epoch": 4.676083779834388,
+      "grad_norm": 1.0402472019195557,
+      "learning_rate": 0.0002,
+      "loss": 1.2146,
+      "step": 4800
+    },
+    {
+      "epoch": 4.685825621042377,
+      "grad_norm": 0.9772472977638245,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 4810
+    },
+    {
+      "epoch": 4.695567462250366,
+      "grad_norm": 0.9082779288291931,
+      "learning_rate": 0.0002,
+      "loss": 1.2667,
+      "step": 4820
+    },
+    {
+      "epoch": 4.705309303458353,
+      "grad_norm": 0.8026862740516663,
+      "learning_rate": 0.0002,
+      "loss": 1.2465,
+      "step": 4830
+    },
+    {
+      "epoch": 4.715051144666342,
+      "grad_norm": 1.1631089448928833,
+      "learning_rate": 0.0002,
+      "loss": 1.3369,
+      "step": 4840
+    },
+    {
+      "epoch": 4.7247929858743305,
+      "grad_norm": 0.9384787678718567,
+      "learning_rate": 0.0002,
+      "loss": 1.261,
+      "step": 4850
+    },
+    {
+      "epoch": 4.734534827082318,
+      "grad_norm": 1.2151581048965454,
+      "learning_rate": 0.0002,
+      "loss": 1.2588,
+      "step": 4860
+    },
+    {
+      "epoch": 4.744276668290307,
+      "grad_norm": 0.9679436087608337,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 4870
+    },
+    {
+      "epoch": 4.754018509498295,
+      "grad_norm": 0.8352158069610596,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 4880
+    },
+    {
+      "epoch": 4.763760350706283,
+      "grad_norm": 1.0205804109573364,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 4890
+    },
+    {
+      "epoch": 4.773502191914272,
+      "grad_norm": 0.9814772605895996,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 4900
+    },
+    {
+      "epoch": 4.78324403312226,
+      "grad_norm": 1.002854347229004,
+      "learning_rate": 0.0002,
+      "loss": 1.3114,
+      "step": 4910
+    },
+    {
+      "epoch": 4.792985874330248,
+      "grad_norm": 1.1609505414962769,
+      "learning_rate": 0.0002,
+      "loss": 1.3143,
+      "step": 4920
+    },
+    {
+      "epoch": 4.802727715538237,
+      "grad_norm": 0.9354982376098633,
+      "learning_rate": 0.0002,
+      "loss": 1.3166,
+      "step": 4930
+    },
+    {
+      "epoch": 4.812469556746225,
+      "grad_norm": 0.9761685729026794,
+      "learning_rate": 0.0002,
+      "loss": 1.2978,
+      "step": 4940
+    },
+    {
+      "epoch": 4.822211397954213,
+      "grad_norm": 1.0604596138000488,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 4950
+    },
+    {
+      "epoch": 4.8319532391622015,
+      "grad_norm": 1.0902808904647827,
+      "learning_rate": 0.0002,
+      "loss": 1.2765,
+      "step": 4960
+    },
+    {
+      "epoch": 4.84169508037019,
+      "grad_norm": 1.0174955129623413,
+      "learning_rate": 0.0002,
+      "loss": 1.3073,
+      "step": 4970
+    },
+    {
+      "epoch": 4.851436921578179,
+      "grad_norm": 1.0995253324508667,
+      "learning_rate": 0.0002,
+      "loss": 1.3141,
+      "step": 4980
+    },
+    {
+      "epoch": 4.8611787627861665,
+      "grad_norm": 0.880993127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3006,
+      "step": 4990
+    },
+    {
+      "epoch": 4.870920603994155,
+      "grad_norm": 0.9472237825393677,
+      "learning_rate": 0.0002,
+      "loss": 1.2547,
+      "step": 5000
+    },
+    {
+      "epoch": 4.880662445202143,
+      "grad_norm": 0.9504236578941345,
+      "learning_rate": 0.0002,
+      "loss": 1.4078,
+      "step": 5010
+    },
+    {
+      "epoch": 4.890404286410131,
+      "grad_norm": 1.1261742115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2791,
+      "step": 5020
+    },
+    {
+      "epoch": 4.90014612761812,
+      "grad_norm": 0.904674768447876,
+      "learning_rate": 0.0002,
+      "loss": 1.3707,
+      "step": 5030
+    },
+    {
+      "epoch": 4.909887968826109,
+      "grad_norm": 0.8828991055488586,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 5040
+    },
+    {
+      "epoch": 4.919629810034096,
+      "grad_norm": 1.0156532526016235,
+      "learning_rate": 0.0002,
+      "loss": 1.2905,
+      "step": 5050
+    },
+    {
+      "epoch": 4.929371651242085,
+      "grad_norm": 0.8975168466567993,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 5060
+    },
+    {
+      "epoch": 4.939113492450073,
+      "grad_norm": 0.9787213802337646,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 5070
+    },
+    {
+      "epoch": 4.948855333658061,
+      "grad_norm": 1.0801568031311035,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 5080
+    },
+    {
+      "epoch": 4.95859717486605,
+      "grad_norm": 1.0655089616775513,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 5090
+    },
+    {
+      "epoch": 4.968339016074038,
+      "grad_norm": 0.8941320180892944,
+      "learning_rate": 0.0002,
+      "loss": 1.2449,
+      "step": 5100
+    },
+    {
+      "epoch": 4.978080857282026,
+      "grad_norm": 1.050621747970581,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 5110
+    },
+    {
+      "epoch": 4.987822698490015,
+      "grad_norm": 0.9724781513214111,
+      "learning_rate": 0.0002,
+      "loss": 1.3791,
+      "step": 5120
+    },
+    {
+      "epoch": 4.997564539698003,
+      "grad_norm": 0.9850538969039917,
+      "learning_rate": 0.0002,
+      "loss": 1.292,
+      "step": 5130
+    },
+    {
+      "epoch": 4.9995129079396,
+      "eval_loss": 2.0824170112609863,
+      "eval_runtime": 55.592,
+      "eval_samples_per_second": 9.12,
+      "eval_steps_per_second": 1.151,
+      "step": 5132
+    },
+    {
+      "epoch": 5.007306380905991,
+      "grad_norm": 1.0096189975738525,
+      "learning_rate": 0.0002,
+      "loss": 1.037,
+      "step": 5140
+    },
+    {
+      "epoch": 5.01704822211398,
+      "grad_norm": 1.2403408288955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0003,
+      "step": 5150
+    },
+    {
+      "epoch": 5.026790063321968,
+      "grad_norm": 1.1243221759796143,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 5160
+    },
+    {
+      "epoch": 5.036531904529956,
+      "grad_norm": 1.4745502471923828,
+      "learning_rate": 0.0002,
+      "loss": 0.9815,
+      "step": 5170
+    },
+    {
+      "epoch": 5.0462737457379445,
+      "grad_norm": 1.1913198232650757,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 5180
+    },
+    {
+      "epoch": 5.056015586945933,
+      "grad_norm": 1.2732855081558228,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 5190
+    },
+    {
+      "epoch": 5.065757428153921,
+      "grad_norm": 1.1737396717071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 5200
+    },
+    {
+      "epoch": 5.075499269361909,
+      "grad_norm": 1.4162768125534058,
+      "learning_rate": 0.0002,
+      "loss": 0.9754,
+      "step": 5210
+    },
+    {
+      "epoch": 5.085241110569898,
+      "grad_norm": 1.528274655342102,
+      "learning_rate": 0.0002,
+      "loss": 1.0333,
+      "step": 5220
+    },
+    {
+      "epoch": 5.094982951777886,
+      "grad_norm": 1.3966618776321411,
+      "learning_rate": 0.0002,
+      "loss": 1.0227,
+      "step": 5230
+    },
+    {
+      "epoch": 5.104724792985874,
+      "grad_norm": 1.3427953720092773,
+      "learning_rate": 0.0002,
+      "loss": 0.987,
+      "step": 5240
+    },
+    {
+      "epoch": 5.114466634193863,
+      "grad_norm": 1.6533905267715454,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 5250
+    },
+    {
+      "epoch": 5.124208475401851,
+      "grad_norm": 1.4114865064620972,
+      "learning_rate": 0.0002,
+      "loss": 1.0452,
+      "step": 5260
+    },
+    {
+      "epoch": 5.133950316609839,
+      "grad_norm": 1.5460708141326904,
+      "learning_rate": 0.0002,
+      "loss": 1.067,
+      "step": 5270
+    },
+    {
+      "epoch": 5.143692157817828,
+      "grad_norm": 1.3491919040679932,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 5280
+    },
+    {
+      "epoch": 5.153433999025816,
+      "grad_norm": 1.2208969593048096,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 5290
+    },
+    {
+      "epoch": 5.163175840233804,
+      "grad_norm": 1.1141403913497925,
+      "learning_rate": 0.0002,
+      "loss": 1.0362,
+      "step": 5300
+    },
+    {
+      "epoch": 5.172917681441793,
+      "grad_norm": 1.2938064336776733,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 5310
+    },
+    {
+      "epoch": 5.1826595226497805,
+      "grad_norm": 1.2704918384552002,
+      "learning_rate": 0.0002,
+      "loss": 1.0438,
+      "step": 5320
+    },
+    {
+      "epoch": 5.192401363857769,
+      "grad_norm": 1.3928544521331787,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 5330
+    },
+    {
+      "epoch": 5.202143205065758,
+      "grad_norm": 1.1993824243545532,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 5340
+    },
+    {
+      "epoch": 5.211885046273745,
+      "grad_norm": 1.5913670063018799,
+      "learning_rate": 0.0002,
+      "loss": 1.0195,
+      "step": 5350
+    },
+    {
+      "epoch": 5.221626887481734,
+      "grad_norm": 1.1577855348587036,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 5360
+    },
+    {
+      "epoch": 5.231368728689723,
+      "grad_norm": 1.4535993337631226,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 5370
+    },
+    {
+      "epoch": 5.24111056989771,
+      "grad_norm": 1.5068976879119873,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 5380
+    },
+    {
+      "epoch": 5.250852411105699,
+      "grad_norm": 1.2365459203720093,
+      "learning_rate": 0.0002,
+      "loss": 1.0068,
+      "step": 5390
+    },
+    {
+      "epoch": 5.2605942523136875,
+      "grad_norm": 1.3197922706604004,
+      "learning_rate": 0.0002,
+      "loss": 1.0145,
+      "step": 5400
+    },
+    {
+      "epoch": 5.270336093521675,
+      "grad_norm": 1.2395117282867432,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 5410
+    },
+    {
+      "epoch": 5.280077934729664,
+      "grad_norm": 1.1841236352920532,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 5420
+    },
+    {
+      "epoch": 5.289819775937652,
+      "grad_norm": 1.218003749847412,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 5430
+    },
+    {
+      "epoch": 5.29956161714564,
+      "grad_norm": 1.2210947275161743,
+      "learning_rate": 0.0002,
+      "loss": 1.0093,
+      "step": 5440
+    },
+    {
+      "epoch": 5.309303458353629,
+      "grad_norm": 1.266006588935852,
+      "learning_rate": 0.0002,
+      "loss": 0.9619,
+      "step": 5450
+    },
+    {
+      "epoch": 5.319045299561617,
+      "grad_norm": 1.2598075866699219,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 5460
+    },
+    {
+      "epoch": 5.328787140769606,
+      "grad_norm": 1.2410019636154175,
+      "learning_rate": 0.0002,
+      "loss": 1.0929,
+      "step": 5470
+    },
+    {
+      "epoch": 5.338528981977594,
+      "grad_norm": 1.249698519706726,
+      "learning_rate": 0.0002,
+      "loss": 1.058,
+      "step": 5480
+    },
+    {
+      "epoch": 5.348270823185582,
+      "grad_norm": 1.2398173809051514,
+      "learning_rate": 0.0002,
+      "loss": 1.0457,
+      "step": 5490
+    },
+    {
+      "epoch": 5.35801266439357,
+      "grad_norm": 1.2416654825210571,
+      "learning_rate": 0.0002,
+      "loss": 1.0139,
+      "step": 5500
+    },
+    {
+      "epoch": 5.3677545056015585,
+      "grad_norm": 1.398706316947937,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 5510
+    },
+    {
+      "epoch": 5.377496346809547,
+      "grad_norm": 1.3049418926239014,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 5520
+    },
+    {
+      "epoch": 5.387238188017536,
+      "grad_norm": 1.2528893947601318,
+      "learning_rate": 0.0002,
+      "loss": 1.0912,
+      "step": 5530
+    },
+    {
+      "epoch": 5.3969800292255234,
+      "grad_norm": 1.2963255643844604,
+      "learning_rate": 0.0002,
+      "loss": 1.0619,
+      "step": 5540
+    },
+    {
+      "epoch": 5.406721870433512,
+      "grad_norm": 1.494231104850769,
+      "learning_rate": 0.0002,
+      "loss": 1.0194,
+      "step": 5550
+    },
+    {
+      "epoch": 5.416463711641501,
+      "grad_norm": 1.2760992050170898,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 5560
+    },
+    {
+      "epoch": 5.426205552849488,
+      "grad_norm": 1.195292592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.1088,
+      "step": 5570
+    },
+    {
+      "epoch": 5.435947394057477,
+      "grad_norm": 1.6408965587615967,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 5580
+    },
+    {
+      "epoch": 5.4456892352654656,
+      "grad_norm": 1.3092058897018433,
+      "learning_rate": 0.0002,
+      "loss": 1.0868,
+      "step": 5590
+    },
+    {
+      "epoch": 5.455431076473453,
+      "grad_norm": 1.2960586547851562,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 5600
+    },
+    {
+      "epoch": 5.465172917681442,
+      "grad_norm": 1.3560487031936646,
+      "learning_rate": 0.0002,
+      "loss": 1.0257,
+      "step": 5610
+    },
+    {
+      "epoch": 5.4749147588894305,
+      "grad_norm": 1.1896311044692993,
+      "learning_rate": 0.0002,
+      "loss": 1.0314,
+      "step": 5620
+    },
+    {
+      "epoch": 5.484656600097418,
+      "grad_norm": 1.3145595788955688,
+      "learning_rate": 0.0002,
+      "loss": 1.0435,
+      "step": 5630
+    },
+    {
+      "epoch": 5.494398441305407,
+      "grad_norm": 1.2207404375076294,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 5640
+    },
+    {
+      "epoch": 5.504140282513395,
+      "grad_norm": 1.266015887260437,
+      "learning_rate": 0.0002,
+      "loss": 1.0823,
+      "step": 5650
+    },
+    {
+      "epoch": 5.513882123721383,
+      "grad_norm": 1.2478289604187012,
+      "learning_rate": 0.0002,
+      "loss": 1.0696,
+      "step": 5660
+    },
+    {
+      "epoch": 5.523623964929372,
+      "grad_norm": 1.4851372241973877,
+      "learning_rate": 0.0002,
+      "loss": 1.0695,
+      "step": 5670
+    },
+    {
+      "epoch": 5.53336580613736,
+      "grad_norm": 1.4478679895401,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5680
+    },
+    {
+      "epoch": 5.543107647345348,
+      "grad_norm": 1.1079537868499756,
+      "learning_rate": 0.0002,
+      "loss": 1.043,
+      "step": 5690
+    },
+    {
+      "epoch": 5.552849488553337,
+      "grad_norm": 1.4201879501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 5700
+    },
+    {
+      "epoch": 5.562591329761325,
+      "grad_norm": 1.2092000246047974,
+      "learning_rate": 0.0002,
+      "loss": 1.0697,
+      "step": 5710
+    },
+    {
+      "epoch": 5.572333170969313,
+      "grad_norm": 1.4515851736068726,
+      "learning_rate": 0.0002,
+      "loss": 0.9868,
+      "step": 5720
+    },
+    {
+      "epoch": 5.5820750121773015,
+      "grad_norm": 1.3260412216186523,
+      "learning_rate": 0.0002,
+      "loss": 1.1547,
+      "step": 5730
+    },
+    {
+      "epoch": 5.59181685338529,
+      "grad_norm": 1.248191475868225,
+      "learning_rate": 0.0002,
+      "loss": 1.1388,
+      "step": 5740
+    },
+    {
+      "epoch": 5.601558694593278,
+      "grad_norm": 1.2037307024002075,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 5750
+    },
+    {
+      "epoch": 5.611300535801266,
+      "grad_norm": 1.341237187385559,
+      "learning_rate": 0.0002,
+      "loss": 1.1425,
+      "step": 5760
+    },
+    {
+      "epoch": 5.621042377009255,
+      "grad_norm": 1.130115270614624,
+      "learning_rate": 0.0002,
+      "loss": 1.0942,
+      "step": 5770
+    },
+    {
+      "epoch": 5.630784218217243,
+      "grad_norm": 1.3834772109985352,
+      "learning_rate": 0.0002,
+      "loss": 1.1029,
+      "step": 5780
+    },
+    {
+      "epoch": 5.640526059425231,
+      "grad_norm": 1.2586270570755005,
+      "learning_rate": 0.0002,
+      "loss": 1.0825,
+      "step": 5790
+    },
+    {
+      "epoch": 5.65026790063322,
+      "grad_norm": 1.3233023881912231,
+      "learning_rate": 0.0002,
+      "loss": 1.0186,
+      "step": 5800
+    },
+    {
+      "epoch": 5.660009741841208,
+      "grad_norm": 1.2711341381072998,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 5810
+    },
+    {
+      "epoch": 5.669751583049196,
+      "grad_norm": 1.3867720365524292,
+      "learning_rate": 0.0002,
+      "loss": 1.0897,
+      "step": 5820
+    },
+    {
+      "epoch": 5.679493424257185,
+      "grad_norm": 1.4783269166946411,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 5830
+    },
+    {
+      "epoch": 5.6892352654651726,
+      "grad_norm": 1.2744768857955933,
+      "learning_rate": 0.0002,
+      "loss": 1.0632,
+      "step": 5840
+    },
+    {
+      "epoch": 5.698977106673161,
+      "grad_norm": 1.3405882120132446,
+      "learning_rate": 0.0002,
+      "loss": 1.1484,
+      "step": 5850
+    },
+    {
+      "epoch": 5.70871894788115,
+      "grad_norm": 1.204300880432129,
+      "learning_rate": 0.0002,
+      "loss": 1.0975,
+      "step": 5860
+    },
+    {
+      "epoch": 5.7184607890891375,
+      "grad_norm": 1.2954572439193726,
+      "learning_rate": 0.0002,
+      "loss": 1.0494,
+      "step": 5870
+    },
+    {
+      "epoch": 5.728202630297126,
+      "grad_norm": 1.5478382110595703,
+      "learning_rate": 0.0002,
+      "loss": 1.0643,
+      "step": 5880
+    },
+    {
+      "epoch": 5.737944471505115,
+      "grad_norm": 1.2095842361450195,
+      "learning_rate": 0.0002,
+      "loss": 1.0582,
+      "step": 5890
+    },
+    {
+      "epoch": 5.747686312713103,
+      "grad_norm": 1.0691519975662231,
+      "learning_rate": 0.0002,
+      "loss": 1.1,
+      "step": 5900
+    },
+    {
+      "epoch": 5.757428153921091,
+      "grad_norm": 1.1920677423477173,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 5910
+    },
+    {
+      "epoch": 5.76716999512908,
+      "grad_norm": 1.2051277160644531,
+      "learning_rate": 0.0002,
+      "loss": 1.1746,
+      "step": 5920
+    },
+    {
+      "epoch": 5.776911836337067,
+      "grad_norm": 1.197490930557251,
+      "learning_rate": 0.0002,
+      "loss": 1.1221,
+      "step": 5930
+    },
+    {
+      "epoch": 5.786653677545056,
+      "grad_norm": 1.2003998756408691,
+      "learning_rate": 0.0002,
+      "loss": 1.07,
+      "step": 5940
+    },
+    {
+      "epoch": 5.7963955187530445,
+      "grad_norm": 1.2323646545410156,
+      "learning_rate": 0.0002,
+      "loss": 1.0938,
+      "step": 5950
+    },
+    {
+      "epoch": 5.806137359961033,
+      "grad_norm": 1.2593932151794434,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 5960
+    },
+    {
+      "epoch": 5.815879201169021,
+      "grad_norm": 1.1835976839065552,
+      "learning_rate": 0.0002,
+      "loss": 1.0829,
+      "step": 5970
+    },
+    {
+      "epoch": 5.825621042377009,
+      "grad_norm": 1.4770104885101318,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 5980
+    },
+    {
+      "epoch": 5.835362883584997,
+      "grad_norm": 1.1025809049606323,
+      "learning_rate": 0.0002,
+      "loss": 1.1934,
+      "step": 5990
+    },
+    {
+      "epoch": 5.845104724792986,
+      "grad_norm": 1.364588975906372,
+      "learning_rate": 0.0002,
+      "loss": 1.1323,
+      "step": 6000
+    },
+    {
+      "epoch": 5.854846566000974,
+      "grad_norm": 1.2340112924575806,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 6010
+    },
+    {
+      "epoch": 5.864588407208963,
+      "grad_norm": 1.4925711154937744,
+      "learning_rate": 0.0002,
+      "loss": 1.1123,
+      "step": 6020
+    },
+    {
+      "epoch": 5.874330248416951,
+      "grad_norm": 1.3516744375228882,
+      "learning_rate": 0.0002,
+      "loss": 1.12,
+      "step": 6030
+    },
+    {
+      "epoch": 5.884072089624939,
+      "grad_norm": 1.2058138847351074,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 6040
+    },
+    {
+      "epoch": 5.893813930832927,
+      "grad_norm": 1.13870108127594,
+      "learning_rate": 0.0002,
+      "loss": 1.1074,
+      "step": 6050
+    },
+    {
+      "epoch": 5.9035557720409155,
+      "grad_norm": 1.1587319374084473,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 6060
+    },
+    {
+      "epoch": 5.913297613248904,
+      "grad_norm": 1.164481520652771,
+      "learning_rate": 0.0002,
+      "loss": 1.1376,
+      "step": 6070
+    },
+    {
+      "epoch": 5.923039454456893,
+      "grad_norm": 1.2115206718444824,
+      "learning_rate": 0.0002,
+      "loss": 1.1262,
+      "step": 6080
+    },
+    {
+      "epoch": 5.93278129566488,
+      "grad_norm": 1.3201590776443481,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 6090
+    },
+    {
+      "epoch": 5.942523136872869,
+      "grad_norm": 1.287380576133728,
+      "learning_rate": 0.0002,
+      "loss": 1.1288,
+      "step": 6100
+    },
+    {
+      "epoch": 5.952264978080858,
+      "grad_norm": 1.1820166110992432,
+      "learning_rate": 0.0002,
+      "loss": 1.1475,
+      "step": 6110
+    },
+    {
+      "epoch": 5.962006819288845,
+      "grad_norm": 1.2550667524337769,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 6120
+    },
+    {
+      "epoch": 5.971748660496834,
+      "grad_norm": 1.3547813892364502,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 6130
+    },
+    {
+      "epoch": 5.9814905017048225,
+      "grad_norm": 1.260842204093933,
+      "learning_rate": 0.0002,
+      "loss": 1.0557,
+      "step": 6140
+    },
+    {
+      "epoch": 5.99123234291281,
+      "grad_norm": 1.1643036603927612,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 6150
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.2628161907196045,
+      "eval_runtime": 57.2379,
+      "eval_samples_per_second": 8.858,
+      "eval_steps_per_second": 1.118,
+      "step": 6159
+    },
+    {
+      "epoch": 6.000974184120799,
+      "grad_norm": 0.9384723901748657,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 6160
+    },
+    {
+      "epoch": 6.0107160253287875,
+      "grad_norm": 2.1525821685791016,
+      "learning_rate": 0.0002,
+      "loss": 0.7335,
+      "step": 6170
+    },
+    {
+      "epoch": 6.020457866536775,
+      "grad_norm": 2.0194077491760254,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 6180
+    },
+    {
+      "epoch": 6.030199707744764,
+      "grad_norm": 1.5257816314697266,
+      "learning_rate": 0.0002,
+      "loss": 0.8443,
+      "step": 6190
+    },
+    {
+      "epoch": 6.039941548952752,
+      "grad_norm": 1.5432662963867188,
+      "learning_rate": 0.0002,
+      "loss": 0.7543,
+      "step": 6200
+    },
+    {
+      "epoch": 6.04968339016074,
+      "grad_norm": 1.6874405145645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8104,
+      "step": 6210
+    },
+    {
+      "epoch": 6.059425231368729,
+      "grad_norm": 1.7346407175064087,
+      "learning_rate": 0.0002,
+      "loss": 0.8395,
+      "step": 6220
+    },
+    {
+      "epoch": 6.069167072576717,
+      "grad_norm": 1.5320781469345093,
+      "learning_rate": 0.0002,
+      "loss": 0.8027,
+      "step": 6230
+    },
+    {
+      "epoch": 6.078908913784705,
+      "grad_norm": 1.4106669425964355,
+      "learning_rate": 0.0002,
+      "loss": 0.7488,
+      "step": 6240
+    },
+    {
+      "epoch": 6.088650754992694,
+      "grad_norm": 1.5568628311157227,
+      "learning_rate": 0.0002,
+      "loss": 0.812,
+      "step": 6250
+    },
+    {
+      "epoch": 6.098392596200682,
+      "grad_norm": 1.6155978441238403,
+      "learning_rate": 0.0002,
+      "loss": 0.8055,
+      "step": 6260
+    },
+    {
+      "epoch": 6.10813443740867,
+      "grad_norm": 1.4820445775985718,
+      "learning_rate": 0.0002,
+      "loss": 0.8225,
+      "step": 6270
+    },
+    {
+      "epoch": 6.1178762786166585,
+      "grad_norm": 1.6163820028305054,
+      "learning_rate": 0.0002,
+      "loss": 0.8599,
+      "step": 6280
+    },
+    {
+      "epoch": 6.127618119824647,
+      "grad_norm": 1.8396387100219727,
+      "learning_rate": 0.0002,
+      "loss": 0.853,
+      "step": 6290
+    },
+    {
+      "epoch": 6.137359961032635,
+      "grad_norm": 1.7181230783462524,
+      "learning_rate": 0.0002,
+      "loss": 0.7768,
+      "step": 6300
+    },
+    {
+      "epoch": 6.147101802240623,
+      "grad_norm": 1.6568509340286255,
+      "learning_rate": 0.0002,
+      "loss": 0.8116,
+      "step": 6310
+    },
+    {
+      "epoch": 6.156843643448612,
+      "grad_norm": 1.3481947183609009,
+      "learning_rate": 0.0002,
+      "loss": 0.8525,
+      "step": 6320
+    },
+    {
+      "epoch": 6.1665854846566,
+      "grad_norm": 1.5788342952728271,
+      "learning_rate": 0.0002,
+      "loss": 0.762,
+      "step": 6330
+    },
+    {
+      "epoch": 6.176327325864588,
+      "grad_norm": 1.5067620277404785,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 6340
+    },
+    {
+      "epoch": 6.186069167072577,
+      "grad_norm": 1.8198208808898926,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 6350
+    },
+    {
+      "epoch": 6.195811008280565,
+      "grad_norm": 1.4012749195098877,
+      "learning_rate": 0.0002,
+      "loss": 0.7867,
+      "step": 6360
+    },
+    {
+      "epoch": 6.205552849488553,
+      "grad_norm": 1.759798288345337,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 6370
+    },
+    {
+      "epoch": 6.215294690696542,
+      "grad_norm": 1.468922734260559,
+      "learning_rate": 0.0002,
+      "loss": 0.7811,
+      "step": 6380
+    },
+    {
+      "epoch": 6.2250365319045295,
+      "grad_norm": 1.3706471920013428,
+      "learning_rate": 0.0002,
+      "loss": 0.8356,
+      "step": 6390
+    },
+    {
+      "epoch": 6.234778373112518,
+      "grad_norm": 1.6397383213043213,
+      "learning_rate": 0.0002,
+      "loss": 0.8096,
+      "step": 6400
+    },
+    {
+      "epoch": 6.244520214320507,
+      "grad_norm": 1.5614187717437744,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 6410
+    },
+    {
+      "epoch": 6.2542620555284945,
+      "grad_norm": 1.7118678092956543,
+      "learning_rate": 0.0002,
+      "loss": 0.8533,
+      "step": 6420
+    },
+    {
+      "epoch": 6.264003896736483,
+      "grad_norm": 1.4041547775268555,
+      "learning_rate": 0.0002,
+      "loss": 0.8653,
+      "step": 6430
+    },
+    {
+      "epoch": 6.273745737944472,
+      "grad_norm": 1.7653605937957764,
+      "learning_rate": 0.0002,
+      "loss": 0.879,
+      "step": 6440
+    },
+    {
+      "epoch": 6.28348757915246,
+      "grad_norm": 2.6219191551208496,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 6450
+    },
+    {
+      "epoch": 6.293229420360448,
+      "grad_norm": 1.4757837057113647,
+      "learning_rate": 0.0002,
+      "loss": 0.8896,
+      "step": 6460
+    },
+    {
+      "epoch": 6.302971261568437,
+      "grad_norm": 1.715598225593567,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 6470
+    },
+    {
+      "epoch": 6.312713102776424,
+      "grad_norm": 1.376216173171997,
+      "learning_rate": 0.0002,
+      "loss": 0.8526,
+      "step": 6480
+    },
+    {
+      "epoch": 6.322454943984413,
+      "grad_norm": 1.7119828462600708,
+      "learning_rate": 0.0002,
+      "loss": 0.8742,
+      "step": 6490
+    },
+    {
+      "epoch": 6.3321967851924015,
+      "grad_norm": 1.4304355382919312,
+      "learning_rate": 0.0002,
+      "loss": 0.7988,
+      "step": 6500
+    },
+    {
+      "epoch": 6.34193862640039,
+      "grad_norm": 1.4889872074127197,
+      "learning_rate": 0.0002,
+      "loss": 0.8539,
+      "step": 6510
+    },
+    {
+      "epoch": 6.351680467608378,
+      "grad_norm": 1.370373010635376,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 6520
+    },
+    {
+      "epoch": 6.361422308816366,
+      "grad_norm": 1.7697709798812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 6530
+    },
+    {
+      "epoch": 6.371164150024355,
+      "grad_norm": 1.495297908782959,
+      "learning_rate": 0.0002,
+      "loss": 0.9421,
+      "step": 6540
+    },
+    {
+      "epoch": 6.380905991232343,
+      "grad_norm": 1.7251347303390503,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 6550
+    },
+    {
+      "epoch": 6.390647832440331,
+      "grad_norm": 1.6909505128860474,
+      "learning_rate": 0.0002,
+      "loss": 0.9327,
+      "step": 6560
+    },
+    {
+      "epoch": 6.40038967364832,
+      "grad_norm": 1.4369314908981323,
+      "learning_rate": 0.0002,
+      "loss": 0.837,
+      "step": 6570
+    },
+    {
+      "epoch": 6.410131514856308,
+      "grad_norm": 1.7803739309310913,
+      "learning_rate": 0.0002,
+      "loss": 0.8572,
+      "step": 6580
+    },
+    {
+      "epoch": 6.419873356064296,
+      "grad_norm": 1.6107097864151,
+      "learning_rate": 0.0002,
+      "loss": 0.9024,
+      "step": 6590
+    },
+    {
+      "epoch": 6.429615197272285,
+      "grad_norm": 1.6151643991470337,
+      "learning_rate": 0.0002,
+      "loss": 0.8469,
+      "step": 6600
+    },
+    {
+      "epoch": 6.4393570384802725,
+      "grad_norm": 1.7159833908081055,
+      "learning_rate": 0.0002,
+      "loss": 0.8791,
+      "step": 6610
+    },
+    {
+      "epoch": 6.449098879688261,
+      "grad_norm": 1.4366064071655273,
+      "learning_rate": 0.0002,
+      "loss": 0.9249,
+      "step": 6620
+    },
+    {
+      "epoch": 6.45884072089625,
+      "grad_norm": 1.6050453186035156,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 6630
+    },
+    {
+      "epoch": 6.468582562104237,
+      "grad_norm": 1.6296740770339966,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 6640
+    },
+    {
+      "epoch": 6.478324403312226,
+      "grad_norm": 1.6181174516677856,
+      "learning_rate": 0.0002,
+      "loss": 0.9228,
+      "step": 6650
+    },
+    {
+      "epoch": 6.488066244520215,
+      "grad_norm": 1.5452176332473755,
+      "learning_rate": 0.0002,
+      "loss": 0.9139,
+      "step": 6660
+    },
+    {
+      "epoch": 6.497808085728202,
+      "grad_norm": 1.3919731378555298,
+      "learning_rate": 0.0002,
+      "loss": 0.9022,
+      "step": 6670
+    },
+    {
+      "epoch": 6.507549926936191,
+      "grad_norm": 1.6456257104873657,
+      "learning_rate": 0.0002,
+      "loss": 0.9046,
+      "step": 6680
+    },
+    {
+      "epoch": 6.5172917681441795,
+      "grad_norm": 1.4147369861602783,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 6690
+    },
+    {
+      "epoch": 6.527033609352167,
+      "grad_norm": 1.7005025148391724,
+      "learning_rate": 0.0002,
+      "loss": 0.8361,
+      "step": 6700
+    },
+    {
+      "epoch": 6.536775450560156,
+      "grad_norm": 1.6032357215881348,
+      "learning_rate": 0.0002,
+      "loss": 0.8738,
+      "step": 6710
+    },
+    {
+      "epoch": 6.5465172917681445,
+      "grad_norm": 1.3454229831695557,
+      "learning_rate": 0.0002,
+      "loss": 0.9796,
+      "step": 6720
+    },
+    {
+      "epoch": 6.556259132976132,
+      "grad_norm": 1.6961418390274048,
+      "learning_rate": 0.0002,
+      "loss": 0.8573,
+      "step": 6730
+    },
+    {
+      "epoch": 6.566000974184121,
+      "grad_norm": 1.78407883644104,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 6740
+    },
+    {
+      "epoch": 6.575742815392109,
+      "grad_norm": 1.6817889213562012,
+      "learning_rate": 0.0002,
+      "loss": 0.8941,
+      "step": 6750
+    },
+    {
+      "epoch": 6.585484656600097,
+      "grad_norm": 1.7894943952560425,
+      "learning_rate": 0.0002,
+      "loss": 0.8765,
+      "step": 6760
+    },
+    {
+      "epoch": 6.595226497808086,
+      "grad_norm": 1.6404837369918823,
+      "learning_rate": 0.0002,
+      "loss": 0.8607,
+      "step": 6770
+    },
+    {
+      "epoch": 6.604968339016074,
+      "grad_norm": 1.5849255323410034,
+      "learning_rate": 0.0002,
+      "loss": 0.8573,
+      "step": 6780
+    },
+    {
+      "epoch": 6.614710180224062,
+      "grad_norm": 1.5993813276290894,
+      "learning_rate": 0.0002,
+      "loss": 0.9575,
+      "step": 6790
+    },
+    {
+      "epoch": 6.624452021432051,
+      "grad_norm": 1.2834863662719727,
+      "learning_rate": 0.0002,
+      "loss": 0.8922,
+      "step": 6800
+    },
+    {
+      "epoch": 6.634193862640039,
+      "grad_norm": 1.7215641736984253,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 6810
+    },
+    {
+      "epoch": 6.643935703848027,
+      "grad_norm": 1.7588146924972534,
+      "learning_rate": 0.0002,
+      "loss": 0.9292,
+      "step": 6820
+    },
+    {
+      "epoch": 6.6536775450560155,
+      "grad_norm": 1.7956023216247559,
+      "learning_rate": 0.0002,
+      "loss": 0.8634,
+      "step": 6830
+    },
+    {
+      "epoch": 6.663419386264004,
+      "grad_norm": 1.5115351676940918,
+      "learning_rate": 0.0002,
+      "loss": 0.8108,
+      "step": 6840
+    },
+    {
+      "epoch": 6.673161227471992,
+      "grad_norm": 1.5660319328308105,
+      "learning_rate": 0.0002,
+      "loss": 0.9329,
+      "step": 6850
+    },
+    {
+      "epoch": 6.68290306867998,
+      "grad_norm": 1.4323679208755493,
+      "learning_rate": 0.0002,
+      "loss": 0.9877,
+      "step": 6860
+    },
+    {
+      "epoch": 6.692644909887969,
+      "grad_norm": 1.662089467048645,
+      "learning_rate": 0.0002,
+      "loss": 0.8732,
+      "step": 6870
+    },
+    {
+      "epoch": 6.702386751095958,
+      "grad_norm": 1.7854869365692139,
+      "learning_rate": 0.0002,
+      "loss": 0.87,
+      "step": 6880
+    },
+    {
+      "epoch": 6.712128592303945,
+      "grad_norm": 1.5491222143173218,
+      "learning_rate": 0.0002,
+      "loss": 0.9105,
+      "step": 6890
+    },
+    {
+      "epoch": 6.721870433511934,
+      "grad_norm": 1.5946987867355347,
+      "learning_rate": 0.0002,
+      "loss": 0.9147,
+      "step": 6900
+    },
+    {
+      "epoch": 6.731612274719922,
+      "grad_norm": 1.6195964813232422,
+      "learning_rate": 0.0002,
+      "loss": 0.9391,
+      "step": 6910
+    },
+    {
+      "epoch": 6.74135411592791,
+      "grad_norm": 1.6366901397705078,
+      "learning_rate": 0.0002,
+      "loss": 0.8947,
+      "step": 6920
+    },
+    {
+      "epoch": 6.751095957135899,
+      "grad_norm": 1.5080382823944092,
+      "learning_rate": 0.0002,
+      "loss": 0.8695,
+      "step": 6930
+    },
+    {
+      "epoch": 6.760837798343887,
+      "grad_norm": 1.742353916168213,
+      "learning_rate": 0.0002,
+      "loss": 0.9124,
+      "step": 6940
+    },
+    {
+      "epoch": 6.770579639551875,
+      "grad_norm": 1.690251111984253,
+      "learning_rate": 0.0002,
+      "loss": 0.9118,
+      "step": 6950
+    },
+    {
+      "epoch": 6.780321480759864,
+      "grad_norm": 1.7103357315063477,
+      "learning_rate": 0.0002,
+      "loss": 0.9039,
+      "step": 6960
+    },
+    {
+      "epoch": 6.7900633219678515,
+      "grad_norm": 1.6630914211273193,
+      "learning_rate": 0.0002,
+      "loss": 0.869,
+      "step": 6970
+    },
+    {
+      "epoch": 6.79980516317584,
+      "grad_norm": 1.423768162727356,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 6980
+    },
+    {
+      "epoch": 6.809547004383829,
+      "grad_norm": 1.7844693660736084,
+      "learning_rate": 0.0002,
+      "loss": 0.9397,
+      "step": 6990
+    },
+    {
+      "epoch": 6.819288845591817,
+      "grad_norm": 1.545282006263733,
+      "learning_rate": 0.0002,
+      "loss": 0.8889,
+      "step": 7000
+    },
+    {
+      "epoch": 6.829030686799805,
+      "grad_norm": 1.4340319633483887,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 7010
+    },
+    {
+      "epoch": 6.838772528007794,
+      "grad_norm": 1.5981626510620117,
+      "learning_rate": 0.0002,
+      "loss": 0.9486,
+      "step": 7020
+    },
+    {
+      "epoch": 6.848514369215782,
+      "grad_norm": 1.5205026865005493,
+      "learning_rate": 0.0002,
+      "loss": 0.9062,
+      "step": 7030
+    },
+    {
+      "epoch": 6.85825621042377,
+      "grad_norm": 1.6999989748001099,
+      "learning_rate": 0.0002,
+      "loss": 0.9245,
+      "step": 7040
+    },
+    {
+      "epoch": 6.8679980516317585,
+      "grad_norm": 1.6392347812652588,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 7050
+    },
+    {
+      "epoch": 6.877739892839747,
+      "grad_norm": 1.637308955192566,
+      "learning_rate": 0.0002,
+      "loss": 0.9275,
+      "step": 7060
+    },
+    {
+      "epoch": 6.887481734047735,
+      "grad_norm": 1.671341896057129,
+      "learning_rate": 0.0002,
+      "loss": 0.9672,
+      "step": 7070
+    },
+    {
+      "epoch": 6.897223575255723,
+      "grad_norm": 1.4437555074691772,
+      "learning_rate": 0.0002,
+      "loss": 0.9726,
+      "step": 7080
+    },
+    {
+      "epoch": 6.906965416463712,
+      "grad_norm": 1.4251935482025146,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 7090
+    },
+    {
+      "epoch": 6.9167072576717,
+      "grad_norm": 1.5106734037399292,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 7100
+    },
+    {
+      "epoch": 6.926449098879688,
+      "grad_norm": 1.670742154121399,
+      "learning_rate": 0.0002,
+      "loss": 0.939,
+      "step": 7110
+    },
+    {
+      "epoch": 6.936190940087677,
+      "grad_norm": 1.4353723526000977,
+      "learning_rate": 0.0002,
+      "loss": 0.8818,
+      "step": 7120
+    },
+    {
+      "epoch": 6.945932781295665,
+      "grad_norm": 1.9437772035598755,
+      "learning_rate": 0.0002,
+      "loss": 0.9354,
+      "step": 7130
+    },
+    {
+      "epoch": 6.955674622503653,
+      "grad_norm": 1.4922038316726685,
+      "learning_rate": 0.0002,
+      "loss": 0.9623,
+      "step": 7140
+    },
+    {
+      "epoch": 6.965416463711642,
+      "grad_norm": 1.489193081855774,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 7150
+    },
+    {
+      "epoch": 6.9751583049196295,
+      "grad_norm": 1.529490351676941,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 7160
+    },
+    {
+      "epoch": 6.984900146127618,
+      "grad_norm": 1.7370105981826782,
+      "learning_rate": 0.0002,
+      "loss": 0.9715,
+      "step": 7170
+    },
+    {
+      "epoch": 6.994641987335607,
+      "grad_norm": 1.5639604330062866,
+      "learning_rate": 0.0002,
+      "loss": 0.921,
+      "step": 7180
+    },
+    {
+      "epoch": 6.9995129079396,
+      "eval_loss": 2.521758794784546,
+      "eval_runtime": 56.1587,
+      "eval_samples_per_second": 9.028,
+      "eval_steps_per_second": 1.14,
+      "step": 7185
+    },
+    {
+      "epoch": 7.004383828543594,
+      "grad_norm": 1.391621470451355,
+      "learning_rate": 0.0002,
+      "loss": 0.8682,
+      "step": 7190
+    },
+    {
+      "epoch": 7.014125669751583,
+      "grad_norm": 2.3696491718292236,
+      "learning_rate": 0.0002,
+      "loss": 0.6591,
+      "step": 7200
+    },
+    {
+      "epoch": 7.023867510959572,
+      "grad_norm": 1.6873828172683716,
+      "learning_rate": 0.0002,
+      "loss": 0.5653,
+      "step": 7210
+    },
+    {
+      "epoch": 7.033609352167559,
+      "grad_norm": 1.8893300294876099,
+      "learning_rate": 0.0002,
+      "loss": 0.6301,
+      "step": 7220
+    },
+    {
+      "epoch": 7.043351193375548,
+      "grad_norm": 1.6323082447052002,
+      "learning_rate": 0.0002,
+      "loss": 0.5653,
+      "step": 7230
+    },
+    {
+      "epoch": 7.0530930345835365,
+      "grad_norm": 1.9979127645492554,
+      "learning_rate": 0.0002,
+      "loss": 0.6402,
+      "step": 7240
+    },
+    {
+      "epoch": 7.062834875791524,
+      "grad_norm": 2.0339183807373047,
+      "learning_rate": 0.0002,
+      "loss": 0.6314,
+      "step": 7250
+    },
+    {
+      "epoch": 7.072576716999513,
+      "grad_norm": 1.6820781230926514,
+      "learning_rate": 0.0002,
+      "loss": 0.615,
+      "step": 7260
+    },
+    {
+      "epoch": 7.0823185582075014,
+      "grad_norm": 2.0400710105895996,
+      "learning_rate": 0.0002,
+      "loss": 0.7023,
+      "step": 7270
+    },
+    {
+      "epoch": 7.092060399415489,
+      "grad_norm": 2.13495135307312,
+      "learning_rate": 0.0002,
+      "loss": 0.6003,
+      "step": 7280
+    },
+    {
+      "epoch": 7.101802240623478,
+      "grad_norm": 1.6993554830551147,
+      "learning_rate": 0.0002,
+      "loss": 0.6243,
+      "step": 7290
+    },
+    {
+      "epoch": 7.111544081831466,
+      "grad_norm": 1.9262464046478271,
+      "learning_rate": 0.0002,
+      "loss": 0.6587,
+      "step": 7300
+    },
+    {
+      "epoch": 7.121285923039454,
+      "grad_norm": 1.8407244682312012,
+      "learning_rate": 0.0002,
+      "loss": 0.6445,
+      "step": 7310
+    },
+    {
+      "epoch": 7.131027764247443,
+      "grad_norm": 1.744294285774231,
+      "learning_rate": 0.0002,
+      "loss": 0.6305,
+      "step": 7320
+    },
+    {
+      "epoch": 7.140769605455431,
+      "grad_norm": 1.7602320909500122,
+      "learning_rate": 0.0002,
+      "loss": 0.6886,
+      "step": 7330
+    },
+    {
+      "epoch": 7.150511446663419,
+      "grad_norm": 1.7360851764678955,
+      "learning_rate": 0.0002,
+      "loss": 0.6582,
+      "step": 7340
+    },
+    {
+      "epoch": 7.160253287871408,
+      "grad_norm": 2.0012850761413574,
+      "learning_rate": 0.0002,
+      "loss": 0.6172,
+      "step": 7350
+    },
+    {
+      "epoch": 7.169995129079396,
+      "grad_norm": 2.064319372177124,
+      "learning_rate": 0.0002,
+      "loss": 0.6636,
+      "step": 7360
+    },
+    {
+      "epoch": 7.179736970287384,
+      "grad_norm": 1.4556169509887695,
+      "learning_rate": 0.0002,
+      "loss": 0.6273,
+      "step": 7370
+    },
+    {
+      "epoch": 7.1894788114953725,
+      "grad_norm": 2.365649938583374,
+      "learning_rate": 0.0002,
+      "loss": 0.6944,
+      "step": 7380
+    },
+    {
+      "epoch": 7.199220652703361,
+      "grad_norm": 1.8271889686584473,
+      "learning_rate": 0.0002,
+      "loss": 0.6572,
+      "step": 7390
+    },
+    {
+      "epoch": 7.208962493911349,
+      "grad_norm": 1.9143747091293335,
+      "learning_rate": 0.0002,
+      "loss": 0.6955,
+      "step": 7400
+    },
+    {
+      "epoch": 7.218704335119337,
+      "grad_norm": 1.5670185089111328,
+      "learning_rate": 0.0002,
+      "loss": 0.6811,
+      "step": 7410
+    },
+    {
+      "epoch": 7.228446176327326,
+      "grad_norm": 1.7452768087387085,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 7420
+    },
+    {
+      "epoch": 7.238188017535315,
+      "grad_norm": 1.7830921411514282,
+      "learning_rate": 0.0002,
+      "loss": 0.6848,
+      "step": 7430
+    },
+    {
+      "epoch": 7.247929858743302,
+      "grad_norm": 1.9281501770019531,
+      "learning_rate": 0.0002,
+      "loss": 0.7018,
+      "step": 7440
+    },
+    {
+      "epoch": 7.257671699951291,
+      "grad_norm": 1.889663815498352,
+      "learning_rate": 0.0002,
+      "loss": 0.6802,
+      "step": 7450
+    },
+    {
+      "epoch": 7.2674135411592795,
+      "grad_norm": 1.704999566078186,
+      "learning_rate": 0.0002,
+      "loss": 0.6457,
+      "step": 7460
+    },
+    {
+      "epoch": 7.277155382367267,
+      "grad_norm": 1.824109435081482,
+      "learning_rate": 0.0002,
+      "loss": 0.6919,
+      "step": 7470
+    },
+    {
+      "epoch": 7.286897223575256,
+      "grad_norm": 1.5378915071487427,
+      "learning_rate": 0.0002,
+      "loss": 0.6814,
+      "step": 7480
+    },
+    {
+      "epoch": 7.296639064783244,
+      "grad_norm": 1.830587387084961,
+      "learning_rate": 0.0002,
+      "loss": 0.7257,
+      "step": 7490
+    },
+    {
+      "epoch": 7.306380905991232,
+      "grad_norm": 2.0029330253601074,
+      "learning_rate": 0.0002,
+      "loss": 0.7036,
+      "step": 7500
+    },
+    {
+      "epoch": 7.316122747199221,
+      "grad_norm": 2.0871448516845703,
+      "learning_rate": 0.0002,
+      "loss": 0.662,
+      "step": 7510
+    },
+    {
+      "epoch": 7.325864588407209,
+      "grad_norm": 1.8416074514389038,
+      "learning_rate": 0.0002,
+      "loss": 0.6856,
+      "step": 7520
+    },
+    {
+      "epoch": 7.335606429615197,
+      "grad_norm": 1.8962771892547607,
+      "learning_rate": 0.0002,
+      "loss": 0.688,
+      "step": 7530
+    },
+    {
+      "epoch": 7.345348270823186,
+      "grad_norm": 1.899487018585205,
+      "learning_rate": 0.0002,
+      "loss": 0.6935,
+      "step": 7540
+    },
+    {
+      "epoch": 7.355090112031174,
+      "grad_norm": 1.8300765752792358,
+      "learning_rate": 0.0002,
+      "loss": 0.667,
+      "step": 7550
+    },
+    {
+      "epoch": 7.364831953239162,
+      "grad_norm": 2.178112268447876,
+      "learning_rate": 0.0002,
+      "loss": 0.7012,
+      "step": 7560
+    },
+    {
+      "epoch": 7.3745737944471506,
+      "grad_norm": 1.8472180366516113,
+      "learning_rate": 0.0002,
+      "loss": 0.6172,
+      "step": 7570
+    },
+    {
+      "epoch": 7.384315635655139,
+      "grad_norm": 1.7787587642669678,
+      "learning_rate": 0.0002,
+      "loss": 0.6944,
+      "step": 7580
+    },
+    {
+      "epoch": 7.394057476863127,
+      "grad_norm": 1.8309564590454102,
+      "learning_rate": 0.0002,
+      "loss": 0.7236,
+      "step": 7590
+    },
+    {
+      "epoch": 7.4037993180711155,
+      "grad_norm": 2.028923273086548,
+      "learning_rate": 0.0002,
+      "loss": 0.7067,
+      "step": 7600
+    },
+    {
+      "epoch": 7.413541159279104,
+      "grad_norm": 1.7393525838851929,
+      "learning_rate": 0.0002,
+      "loss": 0.6656,
+      "step": 7610
+    },
+    {
+      "epoch": 7.423283000487092,
+      "grad_norm": 1.8816498517990112,
+      "learning_rate": 0.0002,
+      "loss": 0.6958,
+      "step": 7620
+    },
+    {
+      "epoch": 7.43302484169508,
+      "grad_norm": 2.4553585052490234,
+      "learning_rate": 0.0002,
+      "loss": 0.6666,
+      "step": 7630
+    },
+    {
+      "epoch": 7.442766682903069,
+      "grad_norm": 1.9045933485031128,
+      "learning_rate": 0.0002,
+      "loss": 0.6964,
+      "step": 7640
+    },
+    {
+      "epoch": 7.452508524111057,
+      "grad_norm": 1.664156198501587,
+      "learning_rate": 0.0002,
+      "loss": 0.6759,
+      "step": 7650
+    },
+    {
+      "epoch": 7.462250365319045,
+      "grad_norm": 1.792748212814331,
+      "learning_rate": 0.0002,
+      "loss": 0.7777,
+      "step": 7660
+    },
+    {
+      "epoch": 7.471992206527034,
+      "grad_norm": 1.8481247425079346,
+      "learning_rate": 0.0002,
+      "loss": 0.6785,
+      "step": 7670
+    },
+    {
+      "epoch": 7.481734047735022,
+      "grad_norm": 2.0541393756866455,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 7680
+    },
+    {
+      "epoch": 7.49147588894301,
+      "grad_norm": 1.594969630241394,
+      "learning_rate": 0.0002,
+      "loss": 0.7126,
+      "step": 7690
+    },
+    {
+      "epoch": 7.501217730150999,
+      "grad_norm": 2.1409924030303955,
+      "learning_rate": 0.0002,
+      "loss": 0.7042,
+      "step": 7700
+    },
+    {
+      "epoch": 7.5109595713589865,
+      "grad_norm": 1.9743319749832153,
+      "learning_rate": 0.0002,
+      "loss": 0.6691,
+      "step": 7710
+    },
+    {
+      "epoch": 7.520701412566975,
+      "grad_norm": 1.866410493850708,
+      "learning_rate": 0.0002,
+      "loss": 0.677,
+      "step": 7720
+    },
+    {
+      "epoch": 7.530443253774964,
+      "grad_norm": 1.9087774753570557,
+      "learning_rate": 0.0002,
+      "loss": 0.7159,
+      "step": 7730
+    },
+    {
+      "epoch": 7.540185094982951,
+      "grad_norm": 1.8624005317687988,
+      "learning_rate": 0.0002,
+      "loss": 0.7564,
+      "step": 7740
+    },
+    {
+      "epoch": 7.54992693619094,
+      "grad_norm": 1.629889726638794,
+      "learning_rate": 0.0002,
+      "loss": 0.7355,
+      "step": 7750
+    },
+    {
+      "epoch": 7.559668777398929,
+      "grad_norm": 2.1364638805389404,
+      "learning_rate": 0.0002,
+      "loss": 0.8001,
+      "step": 7760
+    },
+    {
+      "epoch": 7.569410618606916,
+      "grad_norm": 1.591701865196228,
+      "learning_rate": 0.0002,
+      "loss": 0.7878,
+      "step": 7770
+    },
+    {
+      "epoch": 7.579152459814905,
+      "grad_norm": 2.3200602531433105,
+      "learning_rate": 0.0002,
+      "loss": 0.7484,
+      "step": 7780
+    },
+    {
+      "epoch": 7.5888943010228935,
+      "grad_norm": 1.9998793601989746,
+      "learning_rate": 0.0002,
+      "loss": 0.7407,
+      "step": 7790
+    },
+    {
+      "epoch": 7.598636142230882,
+      "grad_norm": 1.8921900987625122,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 7800
+    },
+    {
+      "epoch": 7.60837798343887,
+      "grad_norm": 1.8826839923858643,
+      "learning_rate": 0.0002,
+      "loss": 0.6898,
+      "step": 7810
+    },
+    {
+      "epoch": 7.618119824646858,
+      "grad_norm": 1.8796452283859253,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 7820
+    },
+    {
+      "epoch": 7.627861665854846,
+      "grad_norm": 1.6528139114379883,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 7830
+    },
+    {
+      "epoch": 7.637603507062835,
+      "grad_norm": 1.9646536111831665,
+      "learning_rate": 0.0002,
+      "loss": 0.7205,
+      "step": 7840
+    },
+    {
+      "epoch": 7.647345348270823,
+      "grad_norm": 1.6951191425323486,
+      "learning_rate": 0.0002,
+      "loss": 0.7276,
+      "step": 7850
+    },
+    {
+      "epoch": 7.657087189478812,
+      "grad_norm": 1.8734302520751953,
+      "learning_rate": 0.0002,
+      "loss": 0.7476,
+      "step": 7860
+    },
+    {
+      "epoch": 7.6668290306868,
+      "grad_norm": 2.140984058380127,
+      "learning_rate": 0.0002,
+      "loss": 0.729,
+      "step": 7870
+    },
+    {
+      "epoch": 7.676570871894788,
+      "grad_norm": 1.8852670192718506,
+      "learning_rate": 0.0002,
+      "loss": 0.7979,
+      "step": 7880
+    },
+    {
+      "epoch": 7.686312713102776,
+      "grad_norm": 2.1172003746032715,
+      "learning_rate": 0.0002,
+      "loss": 0.745,
+      "step": 7890
+    },
+    {
+      "epoch": 7.696054554310765,
+      "grad_norm": 1.8237593173980713,
+      "learning_rate": 0.0002,
+      "loss": 0.7796,
+      "step": 7900
+    },
+    {
+      "epoch": 7.705796395518753,
+      "grad_norm": 2.1399245262145996,
+      "learning_rate": 0.0002,
+      "loss": 0.7203,
+      "step": 7910
+    },
+    {
+      "epoch": 7.715538236726742,
+      "grad_norm": 1.8119547367095947,
+      "learning_rate": 0.0002,
+      "loss": 0.741,
+      "step": 7920
+    },
+    {
+      "epoch": 7.7252800779347295,
+      "grad_norm": 1.943442463874817,
+      "learning_rate": 0.0002,
+      "loss": 0.7826,
+      "step": 7930
+    },
+    {
+      "epoch": 7.735021919142718,
+      "grad_norm": 1.6926734447479248,
+      "learning_rate": 0.0002,
+      "loss": 0.7635,
+      "step": 7940
+    },
+    {
+      "epoch": 7.744763760350706,
+      "grad_norm": 1.6824363470077515,
+      "learning_rate": 0.0002,
+      "loss": 0.7531,
+      "step": 7950
+    },
+    {
+      "epoch": 7.754505601558694,
+      "grad_norm": 1.8615055084228516,
+      "learning_rate": 0.0002,
+      "loss": 0.7203,
+      "step": 7960
+    },
+    {
+      "epoch": 7.764247442766683,
+      "grad_norm": 1.7171595096588135,
+      "learning_rate": 0.0002,
+      "loss": 0.7765,
+      "step": 7970
+    },
+    {
+      "epoch": 7.773989283974672,
+      "grad_norm": 1.9871152639389038,
+      "learning_rate": 0.0002,
+      "loss": 0.7633,
+      "step": 7980
+    },
+    {
+      "epoch": 7.783731125182659,
+      "grad_norm": 1.8975892066955566,
+      "learning_rate": 0.0002,
+      "loss": 0.8113,
+      "step": 7990
+    },
+    {
+      "epoch": 7.793472966390648,
+      "grad_norm": 1.8259385824203491,
+      "learning_rate": 0.0002,
+      "loss": 0.7363,
+      "step": 8000
+    },
+    {
+      "epoch": 7.8032148075986365,
+      "grad_norm": 2.2361183166503906,
+      "learning_rate": 0.0002,
+      "loss": 0.7867,
+      "step": 8010
+    },
+    {
+      "epoch": 7.812956648806624,
+      "grad_norm": 1.64067804813385,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 8020
+    },
+    {
+      "epoch": 7.822698490014613,
+      "grad_norm": 2.0037248134613037,
+      "learning_rate": 0.0002,
+      "loss": 0.7489,
+      "step": 8030
+    },
+    {
+      "epoch": 7.832440331222601,
+      "grad_norm": 1.8022961616516113,
+      "learning_rate": 0.0002,
+      "loss": 0.7219,
+      "step": 8040
+    },
+    {
+      "epoch": 7.842182172430589,
+      "grad_norm": 1.9980754852294922,
+      "learning_rate": 0.0002,
+      "loss": 0.7891,
+      "step": 8050
+    },
+    {
+      "epoch": 7.851924013638578,
+      "grad_norm": 1.632716178894043,
+      "learning_rate": 0.0002,
+      "loss": 0.735,
+      "step": 8060
+    },
+    {
+      "epoch": 7.861665854846566,
+      "grad_norm": 1.6348111629486084,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 8070
+    },
+    {
+      "epoch": 7.871407696054554,
+      "grad_norm": 1.968295693397522,
+      "learning_rate": 0.0002,
+      "loss": 0.7585,
+      "step": 8080
+    },
+    {
+      "epoch": 7.881149537262543,
+      "grad_norm": 1.6947685480117798,
+      "learning_rate": 0.0002,
+      "loss": 0.7531,
+      "step": 8090
+    },
+    {
+      "epoch": 7.890891378470531,
+      "grad_norm": 6.1600341796875,
+      "learning_rate": 0.0002,
+      "loss": 0.7828,
+      "step": 8100
+    },
+    {
+      "epoch": 7.900633219678519,
+      "grad_norm": 1.9334033727645874,
+      "learning_rate": 0.0002,
+      "loss": 0.7612,
+      "step": 8110
+    },
+    {
+      "epoch": 7.9103750608865075,
+      "grad_norm": 1.729058027267456,
+      "learning_rate": 0.0002,
+      "loss": 0.7877,
+      "step": 8120
+    },
+    {
+      "epoch": 7.920116902094496,
+      "grad_norm": 1.8671422004699707,
+      "learning_rate": 0.0002,
+      "loss": 0.788,
+      "step": 8130
+    },
+    {
+      "epoch": 7.929858743302484,
+      "grad_norm": 1.9794875383377075,
+      "learning_rate": 0.0002,
+      "loss": 0.7798,
+      "step": 8140
+    },
+    {
+      "epoch": 7.9396005845104725,
+      "grad_norm": 1.812229037284851,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 8150
+    },
+    {
+      "epoch": 7.949342425718461,
+      "grad_norm": 1.7354048490524292,
+      "learning_rate": 0.0002,
+      "loss": 0.8335,
+      "step": 8160
+    },
+    {
+      "epoch": 7.959084266926449,
+      "grad_norm": 1.7386713027954102,
+      "learning_rate": 0.0002,
+      "loss": 0.7849,
+      "step": 8170
+    },
+    {
+      "epoch": 7.968826108134437,
+      "grad_norm": 1.917111873626709,
+      "learning_rate": 0.0002,
+      "loss": 0.7829,
+      "step": 8180
+    },
+    {
+      "epoch": 7.978567949342426,
+      "grad_norm": 1.7007793188095093,
+      "learning_rate": 0.0002,
+      "loss": 0.7672,
+      "step": 8190
+    },
+    {
+      "epoch": 7.988309790550414,
+      "grad_norm": 1.8241386413574219,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 8200
+    },
+    {
+      "epoch": 7.996103263516805,
+      "eval_loss": 2.877988576889038,
+      "eval_runtime": 57.2196,
+      "eval_samples_per_second": 8.861,
+      "eval_steps_per_second": 1.118,
+      "step": 8208
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8208,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.215936241159373e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-8208/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/special_tokens_map.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer.model b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer_config.json b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-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_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_args.bin b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f9dd281c79a14a5c823e817e96ff2fd5ff07e747
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e8ef519de92b2db62801feef14bb344bb8aa4b619a17c82a1cf2d14a9cfb5a
+size 5560
diff --git a/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_log.jsonl b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..df706e3971e5a2da92eb2bc4eb21dc663e73817a
--- /dev/null
+++ b/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/training_log.jsonl
@@ -0,0 +1,14 @@
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 1643.7049367427826, "total_accumulated_duration": 1643.7049367427826, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0867, "grad_norm": 0.9572330713272095, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.5036, "grad_norm": 0.48326826095581055, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2889, "grad_norm": 0.518182098865509, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0426, "grad_norm": 0.7774201035499573, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9571, "grad_norm": 0.48625442385673523, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.9898, "grad_norm": 0.73747318983078, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.985, "grad_norm": 1.0816543102264404, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9356, "grad_norm": 0.5395983457565308, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9234, "grad_norm": 0.4758393466472626, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8904, "grad_norm": 0.391179621219635, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9042, "grad_norm": 0.41491082310676575, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8507, "grad_norm": 0.3800058364868164, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.8988, "grad_norm": 0.3326554596424103, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8453, "grad_norm": 0.3734934329986572, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7938, "grad_norm": 0.97270667552948, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8532, "grad_norm": 0.3579750657081604, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8489, "grad_norm": 0.474427193403244, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8452, "grad_norm": 0.4048471748828888, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8921, "grad_norm": 0.3494202792644501, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7686, "grad_norm": 0.3681524991989136, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8426, "grad_norm": 0.4028656482696533, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8826, "grad_norm": 0.3967253267765045, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7559, "grad_norm": 0.3345787823200226, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.8261, "grad_norm": 0.3374127149581909, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8527, "grad_norm": 0.31277430057525635, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7443, "grad_norm": 0.35254907608032227, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8618, "grad_norm": 0.34836333990097046, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8531, "grad_norm": 0.37661314010620117, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8427, "grad_norm": 0.3196435570716858, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.8193, "grad_norm": 0.33571064472198486, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7663, "grad_norm": 0.3384968638420105, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8286, "grad_norm": 0.3517991602420807, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8181, "grad_norm": 0.32833170890808105, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8225, "grad_norm": 0.335313618183136, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9181, "grad_norm": 0.35762909054756165, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7967, "grad_norm": 0.3138239085674286, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8608, "grad_norm": 0.36576879024505615, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8007, "grad_norm": 0.3099065124988556, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9198, "grad_norm": 0.30725058913230896, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7981, "grad_norm": 0.32897332310676575, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8976, "grad_norm": 0.3343771994113922, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8105, "grad_norm": 0.4490242302417755, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8567, "grad_norm": 0.33699190616607666, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8487, "grad_norm": 0.3511468768119812, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8524, "grad_norm": 0.313871830701828, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8469, "grad_norm": 0.4389854967594147, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7898, "grad_norm": 0.34989842772483826, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7562, "grad_norm": 0.4221613109111786, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7859, "grad_norm": 0.44692835211753845, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8261, "grad_norm": 0.3849453330039978, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8033, "grad_norm": 0.3177783191204071, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6953, "grad_norm": 0.31056079268455505, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.827, "grad_norm": 0.3379001021385193, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7737, "grad_norm": 0.3038215637207031, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.785, "grad_norm": 0.298612117767334, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.823, "grad_norm": 0.3020114004611969, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.836, "grad_norm": 0.3560065031051636, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8287, "grad_norm": 0.3355957865715027, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7066, "grad_norm": 0.44496026635169983, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7669, "grad_norm": 0.33643242716789246, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.836, "grad_norm": 0.3368454873561859, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7292, "grad_norm": 0.34035852551460266, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8323, "grad_norm": 0.35322463512420654, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8072, "grad_norm": 0.3246999979019165, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7479, "grad_norm": 0.32876884937286377, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8684, "grad_norm": 0.28478550910949707, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8436, "grad_norm": 0.2967577278614044, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8472, "grad_norm": 0.30370232462882996, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7885, "grad_norm": 0.32845404744148254, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8463, "grad_norm": 0.3011477589607239, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7829, "grad_norm": 0.2819054126739502, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8117, "grad_norm": 0.2947770059108734, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8057, "grad_norm": 0.3109278976917267, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7735, "grad_norm": 0.2782546877861023, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7394, "grad_norm": 0.3449978232383728, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7618, "grad_norm": 0.2844734191894531, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8394, "grad_norm": 0.32151708006858826, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8188, "grad_norm": 0.3154081106185913, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8075, "grad_norm": 0.27891838550567627, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.31346985697746277, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7947, "grad_norm": 0.2725819945335388, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7644, "grad_norm": 0.33911818265914917, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7545, "grad_norm": 0.28316545486450195, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7547, "grad_norm": 0.3096636235713959, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7917, "grad_norm": 0.3493746817111969, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7496, "grad_norm": 0.3246673345565796, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7177, "grad_norm": 0.31515568494796753, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7839, "grad_norm": 0.2828545570373535, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7106, "grad_norm": 0.31354808807373047, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7714, "grad_norm": 0.3020916283130646, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7234, "grad_norm": 0.32516390085220337, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7829, "grad_norm": 0.3393784165382385, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8264, "grad_norm": 0.36150357127189636, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8358, "grad_norm": 0.31706321239471436, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8244, "grad_norm": 0.33156445622444153, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8231, "grad_norm": 0.3012140989303589, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7741, "grad_norm": 0.35591211915016174, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8081, "grad_norm": 0.3784671425819397, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8921, "grad_norm": 0.3150336444377899, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8511, "grad_norm": 0.31587284803390503, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.8018, "grad_norm": 0.3279566466808319, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7967, "grad_norm": 0.29044675827026367, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 8614.675071716309, "total_accumulated_duration": 8614.675071716309, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0849, "grad_norm": 0.7493793964385986, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4869, "grad_norm": 0.4989018440246582, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2871, "grad_norm": 0.5077806711196899, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0476, "grad_norm": 0.6642375588417053, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9592, "grad_norm": 0.45972347259521484, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.9894, "grad_norm": 0.4688694179058075, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9818, "grad_norm": 0.5043684840202332, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9287, "grad_norm": 0.4621885418891907, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9208, "grad_norm": 0.42979180812835693, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8931, "grad_norm": 0.4064626395702362, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9035, "grad_norm": 0.40199074149131775, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8504, "grad_norm": 0.44314295053482056, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3227830231189728, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8483, "grad_norm": 0.4403241276741028, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7946, "grad_norm": 0.35792240500450134, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8478, "grad_norm": 0.3626907765865326, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8466, "grad_norm": 0.43099069595336914, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8434, "grad_norm": 0.3468930423259735, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8898, "grad_norm": 0.36435553431510925, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7698, "grad_norm": 0.45654407143592834, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8448, "grad_norm": 0.3559560179710388, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8808, "grad_norm": 0.42246749997138977, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7555, "grad_norm": 0.35775721073150635, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.8263, "grad_norm": 0.3552558422088623, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8525, "grad_norm": 0.33366483449935913, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7423, "grad_norm": 0.3529402017593384, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8591, "grad_norm": 0.35132497549057007, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8538, "grad_norm": 0.35935965180397034, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8453, "grad_norm": 0.32442131638526917, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.3304824233055115, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7609, "grad_norm": 0.3343234956264496, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8234, "grad_norm": 0.42407166957855225, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8189, "grad_norm": 0.3250719904899597, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8235, "grad_norm": 0.3359096348285675, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9144, "grad_norm": 0.36697816848754883, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7934, "grad_norm": 0.3180158734321594, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8631, "grad_norm": 0.3829543888568878, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8041, "grad_norm": 0.3221031725406647, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9187, "grad_norm": 0.31730103492736816, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7975, "grad_norm": 0.3535255789756775, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8966, "grad_norm": 0.35524800419807434, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8135, "grad_norm": 0.442055344581604, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8574, "grad_norm": 0.3454146087169647, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8531, "grad_norm": 0.33116260170936584, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8503, "grad_norm": 0.33906036615371704, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8453, "grad_norm": 0.35210177302360535, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7888, "grad_norm": 0.3600303530693054, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3073132634162903, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7825, "grad_norm": 0.4066285490989685, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8186, "grad_norm": 0.7604546546936035, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8051, "grad_norm": 0.30416470766067505, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6934, "grad_norm": 0.31779664754867554, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8255, "grad_norm": 0.32126376032829285, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7717, "grad_norm": 0.30189672112464905, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7785, "grad_norm": 0.28167518973350525, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8229, "grad_norm": 0.3048870265483856, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8348, "grad_norm": 0.3255627453327179, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8314, "grad_norm": 0.3127131462097168, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7036, "grad_norm": 0.3731038570404053, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7655, "grad_norm": 0.34647998213768005, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8374, "grad_norm": 0.3332764208316803, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7265, "grad_norm": 0.3403870463371277, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8279, "grad_norm": 0.3102012574672699, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8047, "grad_norm": 0.3112141489982605, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7442, "grad_norm": 0.35278451442718506, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8633, "grad_norm": 0.27916669845581055, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.841, "grad_norm": 0.29995933175086975, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8432, "grad_norm": 0.30316805839538574, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7848, "grad_norm": 0.31463387608528137, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8466, "grad_norm": 0.2990514934062958, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7844, "grad_norm": 0.2812942564487457, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8101, "grad_norm": 0.3082747459411621, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8038, "grad_norm": 0.29814761877059937, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7691, "grad_norm": 0.2816784381866455, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.739, "grad_norm": 0.3230450749397278, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7612, "grad_norm": 0.28781887888908386, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.837, "grad_norm": 0.2953524887561798, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8213, "grad_norm": 0.3054346442222595, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.808, "grad_norm": 0.2908236086368561, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.31755733489990234, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7927, "grad_norm": 0.26710301637649536, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7626, "grad_norm": 0.3451886773109436, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7533, "grad_norm": 0.2959114611148834, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7551, "grad_norm": 0.31711632013320923, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7905, "grad_norm": 0.34547725319862366, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7484, "grad_norm": 0.32684141397476196, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7154, "grad_norm": 0.3261931538581848, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7823, "grad_norm": 0.28676143288612366, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7099, "grad_norm": 0.3189515173435211, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.772, "grad_norm": 0.30141714215278625, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7233, "grad_norm": 0.336100310087204, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.781, "grad_norm": 0.3328329026699066, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8258, "grad_norm": 0.31725209951400757, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8332, "grad_norm": 0.3101555407047272, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8226, "grad_norm": 0.3060206174850464, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8227, "grad_norm": 0.289568156003952, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7728, "grad_norm": 0.3252294957637787, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8063, "grad_norm": 0.3643883168697357, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8899, "grad_norm": 0.313032329082489, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8478, "grad_norm": 0.299296498298645, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.8003, "grad_norm": 0.32191869616508484, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7958, "grad_norm": 0.2850085198879242, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 9131.119746685028, "total_accumulated_duration": 9131.119746685028, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0805, "grad_norm": 1.4004040956497192, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.477, "grad_norm": 0.49528834223747253, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2657, "grad_norm": 0.5620803833007812, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0145, "grad_norm": 0.5206792950630188, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9441, "grad_norm": 0.5192418098449707, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.9848, "grad_norm": 0.4760245680809021, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9792, "grad_norm": 0.49701640009880066, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9248, "grad_norm": 0.426954060792923, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9176, "grad_norm": 0.41753074526786804, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8882, "grad_norm": 0.43176209926605225, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9034, "grad_norm": 0.4172838032245636, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8503, "grad_norm": 0.37853118777275085, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.8993, "grad_norm": 0.3643597662448883, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8498, "grad_norm": 0.3552045226097107, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.793, "grad_norm": 0.3611885905265808, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8489, "grad_norm": 0.35341012477874756, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8461, "grad_norm": 0.411731481552124, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8408, "grad_norm": 0.3752767741680145, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8867, "grad_norm": 0.3459947109222412, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7685, "grad_norm": 0.3611522316932678, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8399, "grad_norm": 0.3513851463794708, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8785, "grad_norm": 0.37236350774765015, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7563, "grad_norm": 0.34816354513168335, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.8209, "grad_norm": 0.3423711955547333, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8506, "grad_norm": 0.3034679889678955, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7423, "grad_norm": 0.3754836320877075, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8595, "grad_norm": 0.3404528498649597, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8526, "grad_norm": 0.3615986108779907, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8456, "grad_norm": 0.3140770494937897, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.8208, "grad_norm": 0.32112881541252136, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7583, "grad_norm": 0.34159770607948303, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8255, "grad_norm": 0.9457993507385254, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8174, "grad_norm": 0.3212255537509918, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8235, "grad_norm": 0.34061118960380554, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9153, "grad_norm": 0.3667244017124176, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7942, "grad_norm": 0.3188079595565796, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8638, "grad_norm": 0.35733574628829956, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.7988, "grad_norm": 0.30891889333724976, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.921, "grad_norm": 0.29940274357795715, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7979, "grad_norm": 0.31464695930480957, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8985, "grad_norm": 0.3302725851535797, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8103, "grad_norm": 0.39097800850868225, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8554, "grad_norm": 0.3535297214984894, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8539, "grad_norm": 0.3308466672897339, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8517, "grad_norm": 0.3058496415615082, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8491, "grad_norm": 0.7465469241142273, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7907, "grad_norm": 0.3629920482635498, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7578, "grad_norm": 0.324266254901886, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7821, "grad_norm": 0.3932960331439972, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8214, "grad_norm": 0.3609100580215454, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8025, "grad_norm": 0.29935353994369507, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6927, "grad_norm": 0.3104528784751892, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8232, "grad_norm": 0.32785764336586, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.772, "grad_norm": 0.3005681037902832, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7782, "grad_norm": 0.2888568341732025, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8191, "grad_norm": 0.32374313473701477, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8334, "grad_norm": 0.3289007842540741, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.831, "grad_norm": 0.3174697160720825, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7025, "grad_norm": 0.34157034754753113, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7634, "grad_norm": 0.33474043011665344, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8317, "grad_norm": 0.3204871118068695, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7249, "grad_norm": 0.3411032259464264, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.832, "grad_norm": 0.33133742213249207, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8068, "grad_norm": 0.32993486523628235, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7421, "grad_norm": 0.3632366359233856, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8629, "grad_norm": 0.2831084132194519, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8406, "grad_norm": 0.3102332353591919, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8444, "grad_norm": 0.2975476086139679, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7871, "grad_norm": 0.32190340757369995, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.544466495513916, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7835, "grad_norm": 0.2760232388973236, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8107, "grad_norm": 0.29771652817726135, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8015, "grad_norm": 0.30051174759864807, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7713, "grad_norm": 0.2826128900051117, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7416, "grad_norm": 0.335135281085968, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.763, "grad_norm": 0.28276902437210083, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8398, "grad_norm": 0.29236480593681335, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8163, "grad_norm": 0.28522199392318726, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8069, "grad_norm": 0.28323712944984436, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7938, "grad_norm": 0.30547669529914856, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7934, "grad_norm": 0.27654391527175903, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7623, "grad_norm": 0.3366940915584564, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.753, "grad_norm": 0.2769072353839874, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7555, "grad_norm": 0.32644131779670715, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7931, "grad_norm": 0.32108405232429504, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7489, "grad_norm": 0.3608355224132538, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.3408980667591095, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.782, "grad_norm": 0.2861264944076538, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7092, "grad_norm": 0.3119470477104187, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7732, "grad_norm": 0.35495609045028687, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.724, "grad_norm": 0.3283773362636566, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7782, "grad_norm": 0.34948012232780457, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8235, "grad_norm": 0.31886062026023865, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8344, "grad_norm": 0.31308072805404663, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8224, "grad_norm": 0.3056151866912842, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8239, "grad_norm": 0.28585341572761536, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7766, "grad_norm": 0.31313595175743103, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8047, "grad_norm": 0.3548518121242523, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8885, "grad_norm": 0.3156449496746063, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8459, "grad_norm": 0.30134329199790955, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7983, "grad_norm": 0.3213663101196289, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7925, "grad_norm": 0.2871834337711334, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 7805.627238512039, "total_accumulated_duration": 7805.627238512039, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0792, "grad_norm": 0.7775738835334778, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4866, "grad_norm": 0.5583823323249817, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.265, "grad_norm": 0.5392235517501831, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0227, "grad_norm": 0.5747509598731995, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.947, "grad_norm": 0.43950968980789185, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.9836, "grad_norm": 0.46261805295944214, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9802, "grad_norm": 0.4872327446937561, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9254, "grad_norm": 0.4279601573944092, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9145, "grad_norm": 0.40902411937713623, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8922, "grad_norm": 0.3914395868778229, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9064, "grad_norm": 0.4119057059288025, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8464, "grad_norm": 0.36096957325935364, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.8976, "grad_norm": 0.3246411681175232, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8445, "grad_norm": 0.3493919372558594, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7908, "grad_norm": 0.38220298290252686, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.848, "grad_norm": 0.35742080211639404, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.846, "grad_norm": 0.8206163048744202, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8433, "grad_norm": 0.41155320405960083, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8887, "grad_norm": 0.36376968026161194, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7691, "grad_norm": 0.3520759046077728, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8431, "grad_norm": 0.3779120147228241, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8806, "grad_norm": 0.36827215552330017, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7555, "grad_norm": 0.35667774081230164, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.8234, "grad_norm": 0.3537030518054962, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8506, "grad_norm": 0.3095904290676117, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.74, "grad_norm": 0.3664686679840088, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8595, "grad_norm": 0.3403360843658447, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8514, "grad_norm": 0.35468989610671997, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.846, "grad_norm": 0.33379730582237244, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.8197, "grad_norm": 0.3371599614620209, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7603, "grad_norm": 0.345524400472641, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8263, "grad_norm": 0.33196690678596497, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8161, "grad_norm": 0.33003339171409607, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8261, "grad_norm": 0.34377405047416687, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9168, "grad_norm": 0.38273748755455017, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7992, "grad_norm": 0.31873393058776855, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8673, "grad_norm": 0.4106711149215698, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.802, "grad_norm": 0.3180122673511505, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9229, "grad_norm": 0.30043068528175354, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7983, "grad_norm": 0.3154022991657257, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.9015, "grad_norm": 0.3352239727973938, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8102, "grad_norm": 0.44572168588638306, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8588, "grad_norm": 0.3307209312915802, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8497, "grad_norm": 0.3264922499656677, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8515, "grad_norm": 0.3088473081588745, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8498, "grad_norm": 0.9683569073677063, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7902, "grad_norm": 0.38158977031707764, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7569, "grad_norm": 0.3046931326389313, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7835, "grad_norm": 0.3930199444293976, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8243, "grad_norm": 0.4819939136505127, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8042, "grad_norm": 0.30577245354652405, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6942, "grad_norm": 0.3192637264728546, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8252, "grad_norm": 0.3385009467601776, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7762, "grad_norm": 0.3092128336429596, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7805, "grad_norm": 0.29258453845977783, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8223, "grad_norm": 0.30027705430984497, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8365, "grad_norm": 0.3087688982486725, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8318, "grad_norm": 0.6585562229156494, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7023, "grad_norm": 0.34260162711143494, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7602, "grad_norm": 0.3322044909000397, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.835, "grad_norm": 0.3490358591079712, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7277, "grad_norm": 0.33683666586875916, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8276, "grad_norm": 0.33935579657554626, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8063, "grad_norm": 0.3583163917064667, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7456, "grad_norm": 0.3733830153942108, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8643, "grad_norm": 0.2808888256549835, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8403, "grad_norm": 0.29853978753089905, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8423, "grad_norm": 0.2974698841571808, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7887, "grad_norm": 0.31382060050964355, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8474, "grad_norm": 0.3384411334991455, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7824, "grad_norm": 0.279596209526062, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8112, "grad_norm": 0.2944697141647339, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8058, "grad_norm": 0.3064538538455963, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7707, "grad_norm": 0.28189340233802795, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3372725546360016, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7594, "grad_norm": 0.2781224250793457, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.836, "grad_norm": 0.2901884913444519, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8183, "grad_norm": 0.2983466386795044, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8083, "grad_norm": 0.28513798117637634, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.795, "grad_norm": 0.32360851764678955, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7908, "grad_norm": 0.274665892124176, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7632, "grad_norm": 0.3458949029445648, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7529, "grad_norm": 0.2805227041244507, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31838613748550415, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7912, "grad_norm": 0.3196384012699127, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7497, "grad_norm": 0.3167778253555298, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7159, "grad_norm": 0.3295934498310089, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7826, "grad_norm": 0.2816080152988434, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7086, "grad_norm": 0.3783347010612488, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.773, "grad_norm": 0.31769317388534546, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7215, "grad_norm": 0.3217010796070099, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7808, "grad_norm": 0.44784948229789734, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8243, "grad_norm": 0.311459481716156, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.835, "grad_norm": 0.31605151295661926, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8204, "grad_norm": 0.2969941794872284, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8237, "grad_norm": 0.29052597284317017, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7735, "grad_norm": 0.5308746099472046, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8038, "grad_norm": 0.3567570745944977, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8888, "grad_norm": 0.3252796232700348, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8484, "grad_norm": 0.4144253432750702, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.798, "grad_norm": 0.3273269236087799, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7948, "grad_norm": 0.2791995108127594, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 5217.032011985779, "total_accumulated_duration": 5217.032011985779, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0768, "grad_norm": 0.7021015882492065, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4631, "grad_norm": 0.5293550491333008, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2465, "grad_norm": 0.7597886323928833, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0107, "grad_norm": 0.5716570019721985, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9413, "grad_norm": 0.42367565631866455, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.9828, "grad_norm": 0.4324796199798584, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.981, "grad_norm": 0.5050645470619202, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9234, "grad_norm": 0.42312121391296387, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9152, "grad_norm": 0.41372519731521606, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8904, "grad_norm": 0.3871922194957733, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9072, "grad_norm": 0.40193668007850647, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8457, "grad_norm": 0.3660767078399658, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.8967, "grad_norm": 0.31849604845046997, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8465, "grad_norm": 0.34399211406707764, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7901, "grad_norm": 0.32782959938049316, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8499, "grad_norm": 0.33730486035346985, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8507, "grad_norm": 0.47183865308761597, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8449, "grad_norm": 0.41224056482315063, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8872, "grad_norm": 0.3493252098560333, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7698, "grad_norm": 0.3443562686443329, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8442, "grad_norm": 0.3719293177127838, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8795, "grad_norm": 0.37245652079582214, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7547, "grad_norm": 0.3301680386066437, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.8252, "grad_norm": 0.36012348532676697, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8539, "grad_norm": 0.43787631392478943, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7411, "grad_norm": 0.40697792172431946, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8604, "grad_norm": 0.37754276394844055, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8508, "grad_norm": 0.37219709157943726, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8499, "grad_norm": 0.32134678959846497, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.8204, "grad_norm": 0.323728084564209, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7651, "grad_norm": 0.3529917895793915, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8289, "grad_norm": 0.3383786976337433, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8153, "grad_norm": 0.32772132754325867, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8242, "grad_norm": 0.337752103805542, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9168, "grad_norm": 0.35874703526496887, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.8002, "grad_norm": 0.32408469915390015, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8643, "grad_norm": 0.35104885697364807, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8018, "grad_norm": 0.30929499864578247, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9232, "grad_norm": 0.3031355142593384, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7957, "grad_norm": 0.3190555274486542, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8983, "grad_norm": 0.44238677620887756, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8108, "grad_norm": 0.3833184838294983, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8567, "grad_norm": 0.3284796476364136, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8518, "grad_norm": 0.337785929441452, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8543, "grad_norm": 0.3230440020561218, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8484, "grad_norm": 0.4421052634716034, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7897, "grad_norm": 0.35508641600608826, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7569, "grad_norm": 0.2951701879501343, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7841, "grad_norm": 0.4701220393180847, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8226, "grad_norm": 0.3613520860671997, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8047, "grad_norm": 0.302544504404068, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6907, "grad_norm": 0.3328293561935425, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8228, "grad_norm": 0.3272629976272583, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7731, "grad_norm": 0.31455153226852417, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.78, "grad_norm": 0.2919689118862152, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.30982503294944763, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8361, "grad_norm": 0.31483978033065796, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8301, "grad_norm": 0.33073070645332336, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7029, "grad_norm": 0.34696322679519653, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7615, "grad_norm": 0.3365851640701294, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8371, "grad_norm": 0.3475584089756012, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7241, "grad_norm": 0.711076021194458, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.829, "grad_norm": 0.31679967045783997, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8055, "grad_norm": 0.3273877203464508, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7441, "grad_norm": 0.5061460137367249, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8649, "grad_norm": 0.2813282907009125, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8429, "grad_norm": 0.30541664361953735, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8458, "grad_norm": 0.3182958662509918, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7871, "grad_norm": 0.31644582748413086, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8458, "grad_norm": 0.29679274559020996, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7837, "grad_norm": 0.27372097969055176, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8114, "grad_norm": 0.3357728123664856, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8047, "grad_norm": 0.3300599157810211, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7673, "grad_norm": 0.27955976128578186, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7373, "grad_norm": 0.32679328322410583, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7619, "grad_norm": 0.2867962121963501, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8399, "grad_norm": 0.3120456039905548, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8224, "grad_norm": 0.2997247576713562, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8069, "grad_norm": 0.28892984986305237, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7963, "grad_norm": 0.3257523775100708, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7935, "grad_norm": 0.26903730630874634, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7628, "grad_norm": 0.33329081535339355, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7534, "grad_norm": 0.28643161058425903, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7525, "grad_norm": 0.31738072633743286, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7929, "grad_norm": 0.3202503025531769, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7488, "grad_norm": 0.329441100358963, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7152, "grad_norm": 0.33071649074554443, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7853, "grad_norm": 0.29738152027130127, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7115, "grad_norm": 0.3103555738925934, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7728, "grad_norm": 0.3052198588848114, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7216, "grad_norm": 0.31931647658348083, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.782, "grad_norm": 0.33478689193725586, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8271, "grad_norm": 0.32925254106521606, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8361, "grad_norm": 0.34086015820503235, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8234, "grad_norm": 0.3911075294017792, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8241, "grad_norm": 0.28895238041877747, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7749, "grad_norm": 0.3233981728553772, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8016, "grad_norm": 0.35691210627555847, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8894, "grad_norm": 0.31073638796806335, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8502, "grad_norm": 0.2948456108570099, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7993, "grad_norm": 0.3256973326206207, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7939, "grad_norm": 0.2843184173107147, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 5042.019628286362, "total_accumulated_duration": 5042.019628286362, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0786, "grad_norm": 0.6911329030990601, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4641, "grad_norm": 0.5412330627441406, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.267, "grad_norm": 0.583145022392273, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.024, "grad_norm": 0.5168295502662659, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9514, "grad_norm": 0.4544302821159363, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.989, "grad_norm": 0.45545899868011475, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9795, "grad_norm": 0.47717148065567017, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9267, "grad_norm": 0.43077918887138367, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9174, "grad_norm": 0.41169410943984985, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8914, "grad_norm": 0.4109228253364563, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9063, "grad_norm": 0.727739155292511, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.849, "grad_norm": 0.39693373441696167, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.8961, "grad_norm": 0.32479676604270935, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.8476, "grad_norm": 0.41577887535095215, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7912, "grad_norm": 0.3692895770072937, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8483, "grad_norm": 0.3582785427570343, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8469, "grad_norm": 0.43169599771499634, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.8429, "grad_norm": 0.38158753514289856, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8899, "grad_norm": 0.3516024053096771, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7692, "grad_norm": 0.4199542701244354, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8455, "grad_norm": 0.35992029309272766, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8784, "grad_norm": 0.40170818567276, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7562, "grad_norm": 0.34489646553993225, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.825, "grad_norm": 0.36570557951927185, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8508, "grad_norm": 0.3102056682109833, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7412, "grad_norm": 0.34393879771232605, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8589, "grad_norm": 0.3517219126224518, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8518, "grad_norm": 0.3594422936439514, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.32002630829811096, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.8205, "grad_norm": 0.3338747024536133, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7622, "grad_norm": 0.3600769639015198, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8237, "grad_norm": 0.3585824966430664, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8172, "grad_norm": 0.333251029253006, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8226, "grad_norm": 0.347112238407135, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9134, "grad_norm": 0.3596053421497345, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7916, "grad_norm": 0.3617101311683655, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8656, "grad_norm": 0.36017975211143494, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8021, "grad_norm": 0.31422123312950134, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9224, "grad_norm": 0.3083730638027191, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7993, "grad_norm": 0.31650277972221375, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.9, "grad_norm": 0.3626275062561035, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8122, "grad_norm": 0.4097851812839508, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8552, "grad_norm": 0.32672059535980225, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8498, "grad_norm": 0.3282705843448639, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8484, "grad_norm": 0.3159716725349426, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8478, "grad_norm": 0.4060676097869873, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7905, "grad_norm": 0.3285563886165619, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7555, "grad_norm": 0.3229023814201355, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7853, "grad_norm": 0.4205825924873352, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8225, "grad_norm": 0.3491184115409851, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8043, "grad_norm": 0.30142080783843994, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6917, "grad_norm": 0.32370421290397644, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8231, "grad_norm": 0.3260400891304016, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.771, "grad_norm": 0.30405792593955994, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7804, "grad_norm": 0.3368788957595825, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8228, "grad_norm": 0.29851457476615906, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8358, "grad_norm": 0.3424811065196991, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8333, "grad_norm": 0.3058519661426544, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7032, "grad_norm": 0.3714658319950104, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.7629, "grad_norm": 0.33727630972862244, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8341, "grad_norm": 0.32690030336380005, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.727, "grad_norm": 0.33565303683280945, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8283, "grad_norm": 0.3237195909023285, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8053, "grad_norm": 0.3161337375640869, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7407, "grad_norm": 0.7614491581916809, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8636, "grad_norm": 0.2802790105342865, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8411, "grad_norm": 0.330805242061615, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.8439, "grad_norm": 0.2941242456436157, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7857, "grad_norm": 0.32636013627052307, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8445, "grad_norm": 0.3066718876361847, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7808, "grad_norm": 0.2920783460140228, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8099, "grad_norm": 0.2931442856788635, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8038, "grad_norm": 0.314094215631485, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7718, "grad_norm": 0.2800949215888977, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7382, "grad_norm": 0.3249237537384033, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7624, "grad_norm": 0.2794058918952942, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8376, "grad_norm": 0.37049400806427, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8185, "grad_norm": 0.2909463047981262, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8073, "grad_norm": 0.28539058566093445, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7946, "grad_norm": 0.3309869170188904, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7916, "grad_norm": 0.2745763957500458, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7661, "grad_norm": 0.363674521446228, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7543, "grad_norm": 0.2761007249355316, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7543, "grad_norm": 0.5074517130851746, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.7935, "grad_norm": 0.32050713896751404, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.749, "grad_norm": 0.3138967752456665, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.714, "grad_norm": 0.31975120306015015, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7818, "grad_norm": 0.3553573787212372, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7097, "grad_norm": 0.35411757230758667, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7703, "grad_norm": 0.2980964183807373, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7259, "grad_norm": 0.3381917476654053, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7781, "grad_norm": 0.3236033320426941, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8248, "grad_norm": 0.3197905123233795, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8367, "grad_norm": 0.31410661339759827, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.8219, "grad_norm": 0.3096868395805359, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8244, "grad_norm": 0.28688159584999084, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7729, "grad_norm": 0.5106386542320251, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8041, "grad_norm": 0.38583698868751526, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8887, "grad_norm": 0.32006171345710754, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8518, "grad_norm": 0.3037484586238861, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7987, "grad_norm": 0.32314303517341614, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7957, "grad_norm": 0.28476133942604065, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 0.9995129079396006, "step": 1026, "epoch_duration": 1682.2428288459778, "total_accumulated_duration": 1682.2428288459778, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}]}
+{"epoch": 2.0, "step": 2053, "epoch_duration": 2075.387904882431, "total_accumulated_duration": 3757.630733728409, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}]}
+{"epoch": 2.9995129079396006, "step": 3079, "epoch_duration": 2152.3596365451813, "total_accumulated_duration": 5909.99037027359, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}]}
+{"epoch": 4.0, "step": 4106, "epoch_duration": 1664.8996212482452, "total_accumulated_duration": 7574.889991521835, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}, {"eval_loss": 1.8463934659957886, "eval_runtime": 56.2401, "eval_samples_per_second": 9.015, "eval_steps_per_second": 1.138, "epoch": 2.9995129079396006, "step": 3079}, {"loss": 1.5744, "grad_norm": 0.49992576241493225, "learning_rate": 0.0002, "epoch": 3.0004870920603994, "step": 3080}, {"loss": 1.4125, "grad_norm": 0.8242783546447754, "learning_rate": 0.0002, "epoch": 3.0102289332683876, "step": 3090}, {"loss": 1.394, "grad_norm": 0.6330569386482239, "learning_rate": 0.0002, "epoch": 3.019970774476376, "step": 3100}, {"loss": 1.4942, "grad_norm": 0.566097617149353, "learning_rate": 0.0002, "epoch": 3.0297126156843643, "step": 3110}, {"loss": 1.4365, "grad_norm": 0.6337586045265198, "learning_rate": 0.0002, "epoch": 3.0394544568923525, "step": 3120}, {"loss": 1.3916, "grad_norm": 0.7339403033256531, "learning_rate": 0.0002, "epoch": 3.049196298100341, "step": 3130}, {"loss": 1.4617, "grad_norm": 0.7187346816062927, "learning_rate": 0.0002, "epoch": 3.0589381393083293, "step": 3140}, {"loss": 1.3453, "grad_norm": 0.7116255760192871, "learning_rate": 0.0002, "epoch": 3.0686799805163174, "step": 3150}, {"loss": 1.4452, "grad_norm": 0.6493807435035706, "learning_rate": 0.0002, "epoch": 3.078421821724306, "step": 3160}, {"loss": 1.351, "grad_norm": 0.6777266263961792, "learning_rate": 0.0002, "epoch": 3.088163662932294, "step": 3170}, {"loss": 1.4362, "grad_norm": 0.6342006325721741, "learning_rate": 0.0002, "epoch": 3.0979055041402823, "step": 3180}, {"loss": 1.4748, "grad_norm": 0.6608964204788208, "learning_rate": 0.0002, "epoch": 3.107647345348271, "step": 3190}, {"loss": 1.375, "grad_norm": 0.7230247259140015, "learning_rate": 0.0002, "epoch": 3.117389186556259, "step": 3200}, {"loss": 1.4049, "grad_norm": 0.650368332862854, "learning_rate": 0.0002, "epoch": 3.1271310277642472, "step": 3210}, {"loss": 1.409, "grad_norm": 0.7319342494010925, "learning_rate": 0.0002, "epoch": 3.136872868972236, "step": 3220}, {"loss": 1.3872, "grad_norm": 0.7159963846206665, "learning_rate": 0.0002, "epoch": 3.146614710180224, "step": 3230}, {"loss": 1.5076, "grad_norm": 0.8905230164527893, "learning_rate": 0.0002, "epoch": 3.156356551388212, "step": 3240}, {"loss": 1.3161, "grad_norm": 0.6920804381370544, "learning_rate": 0.0002, "epoch": 3.1660983925962007, "step": 3250}, {"loss": 1.3786, "grad_norm": 0.6782063841819763, "learning_rate": 0.0002, "epoch": 3.175840233804189, "step": 3260}, {"loss": 1.5153, "grad_norm": 0.735325276851654, "learning_rate": 0.0002, "epoch": 3.1855820750121775, "step": 3270}, {"loss": 1.4027, "grad_norm": 0.6657978296279907, "learning_rate": 0.0002, "epoch": 3.1953239162201656, "step": 3280}, {"loss": 1.3456, "grad_norm": 0.771315336227417, "learning_rate": 0.0002, "epoch": 3.205065757428154, "step": 3290}, {"loss": 1.3236, "grad_norm": 0.6492983102798462, "learning_rate": 0.0002, "epoch": 3.2148075986361424, "step": 3300}, {"loss": 1.4125, "grad_norm": 0.7513770461082458, "learning_rate": 0.0002, "epoch": 3.2245494398441306, "step": 3310}, {"loss": 1.4032, "grad_norm": 0.7091423869132996, "learning_rate": 0.0002, "epoch": 3.2342912810521187, "step": 3320}, {"loss": 1.4585, "grad_norm": 0.6663975119590759, "learning_rate": 0.0002, "epoch": 3.2440331222601073, "step": 3330}, {"loss": 1.3968, "grad_norm": 0.6813122034072876, "learning_rate": 0.0002, "epoch": 3.2537749634680955, "step": 3340}, {"loss": 1.3681, "grad_norm": 0.6602569818496704, "learning_rate": 0.0002, "epoch": 3.2635168046760836, "step": 3350}, {"loss": 1.4533, "grad_norm": 0.718270480632782, "learning_rate": 0.0002, "epoch": 3.2732586458840722, "step": 3360}, {"loss": 1.4076, "grad_norm": 0.6884173154830933, "learning_rate": 0.0002, "epoch": 3.2830004870920604, "step": 3370}, {"loss": 1.4144, "grad_norm": 0.7039775848388672, "learning_rate": 0.0002, "epoch": 3.2927423283000485, "step": 3380}, {"loss": 1.5077, "grad_norm": 0.7444299459457397, "learning_rate": 0.0002, "epoch": 3.302484169508037, "step": 3390}, {"loss": 1.4255, "grad_norm": 0.7187064290046692, "learning_rate": 0.0002, "epoch": 3.3122260107160253, "step": 3400}, {"loss": 1.3684, "grad_norm": 0.599396288394928, "learning_rate": 0.0002, "epoch": 3.3219678519240134, "step": 3410}, {"loss": 1.4819, "grad_norm": 0.7670390009880066, "learning_rate": 0.0002, "epoch": 3.331709693132002, "step": 3420}, {"loss": 1.4411, "grad_norm": 0.6654478311538696, "learning_rate": 0.0002, "epoch": 3.34145153433999, "step": 3430}, {"loss": 1.4257, "grad_norm": 0.6644385457038879, "learning_rate": 0.0002, "epoch": 3.351193375547979, "step": 3440}, {"loss": 1.4508, "grad_norm": 0.6974098086357117, "learning_rate": 0.0002, "epoch": 3.360935216755967, "step": 3450}, {"loss": 1.3807, "grad_norm": 0.7350399494171143, "learning_rate": 0.0002, "epoch": 3.370677057963955, "step": 3460}, {"loss": 1.4176, "grad_norm": 0.714721143245697, "learning_rate": 0.0002, "epoch": 3.3804188991719437, "step": 3470}, {"loss": 1.4325, "grad_norm": 0.7006027698516846, "learning_rate": 0.0002, "epoch": 3.390160740379932, "step": 3480}, {"loss": 1.4888, "grad_norm": 0.6767925024032593, "learning_rate": 0.0002, "epoch": 3.39990258158792, "step": 3490}, {"loss": 1.4116, "grad_norm": 0.6721355319023132, "learning_rate": 0.0002, "epoch": 3.4096444227959086, "step": 3500}, {"loss": 1.443, "grad_norm": 0.6845725178718567, "learning_rate": 0.0002, "epoch": 3.419386264003897, "step": 3510}, {"loss": 1.4832, "grad_norm": 0.6882196664810181, "learning_rate": 0.0002, "epoch": 3.429128105211885, "step": 3520}, {"loss": 1.4962, "grad_norm": 0.7663240432739258, "learning_rate": 0.0002, "epoch": 3.4388699464198735, "step": 3530}, {"loss": 1.4644, "grad_norm": 0.6304219365119934, "learning_rate": 0.0002, "epoch": 3.4486117876278617, "step": 3540}, {"loss": 1.4918, "grad_norm": 0.668678879737854, "learning_rate": 0.0002, "epoch": 3.45835362883585, "step": 3550}, {"loss": 1.4874, "grad_norm": 0.7526912093162537, "learning_rate": 0.0002, "epoch": 3.4680954700438384, "step": 3560}, {"loss": 1.4249, "grad_norm": 1.089495301246643, "learning_rate": 0.0002, "epoch": 3.4778373112518266, "step": 3570}, {"loss": 1.3871, "grad_norm": 0.7282902002334595, "learning_rate": 0.0002, "epoch": 3.4875791524598148, "step": 3580}, {"loss": 1.5077, "grad_norm": 0.6540156602859497, "learning_rate": 0.0002, "epoch": 3.4973209936678034, "step": 3590}, {"loss": 1.4367, "grad_norm": 0.6449568867683411, "learning_rate": 0.0002, "epoch": 3.5070628348757915, "step": 3600}, {"loss": 1.4532, "grad_norm": 0.7262216210365295, "learning_rate": 0.0002, "epoch": 3.5168046760837797, "step": 3610}, {"loss": 1.4374, "grad_norm": 0.6048615574836731, "learning_rate": 0.0002, "epoch": 3.5265465172917683, "step": 3620}, {"loss": 1.3877, "grad_norm": 0.6780537366867065, "learning_rate": 0.0002, "epoch": 3.5362883584997564, "step": 3630}, {"loss": 1.422, "grad_norm": 0.6851925253868103, "learning_rate": 0.0002, "epoch": 3.5460301997077446, "step": 3640}, {"loss": 1.3425, "grad_norm": 0.6530634164810181, "learning_rate": 0.0002, "epoch": 3.555772040915733, "step": 3650}, {"loss": 1.4879, "grad_norm": 0.7193992733955383, "learning_rate": 0.0002, "epoch": 3.5655138821237213, "step": 3660}, {"loss": 1.4555, "grad_norm": 0.767496645450592, "learning_rate": 0.0002, "epoch": 3.5752557233317095, "step": 3670}, {"loss": 1.4824, "grad_norm": 0.6912919282913208, "learning_rate": 0.0002, "epoch": 3.584997564539698, "step": 3680}, {"loss": 1.4497, "grad_norm": 0.7383436560630798, "learning_rate": 0.0002, "epoch": 3.5947394057476862, "step": 3690}, {"loss": 1.4822, "grad_norm": 0.6746662855148315, "learning_rate": 0.0002, "epoch": 3.6044812469556744, "step": 3700}, {"loss": 1.4904, "grad_norm": 0.6885138750076294, "learning_rate": 0.0002, "epoch": 3.614223088163663, "step": 3710}, {"loss": 1.4044, "grad_norm": 0.6694392561912537, "learning_rate": 0.0002, "epoch": 3.623964929371651, "step": 3720}, {"loss": 1.3719, "grad_norm": 0.812358021736145, "learning_rate": 0.0002, "epoch": 3.6337067705796393, "step": 3730}, {"loss": 1.4603, "grad_norm": 0.7267130017280579, "learning_rate": 0.0002, "epoch": 3.643448611787628, "step": 3740}, {"loss": 1.4574, "grad_norm": 0.6958749294281006, "learning_rate": 0.0002, "epoch": 3.653190452995616, "step": 3750}, {"loss": 1.4346, "grad_norm": 0.6805673241615295, "learning_rate": 0.0002, "epoch": 3.6629322942036042, "step": 3760}, {"loss": 1.4338, "grad_norm": 0.7184410095214844, "learning_rate": 0.0002, "epoch": 3.672674135411593, "step": 3770}, {"loss": 1.3935, "grad_norm": 0.7716330289840698, "learning_rate": 0.0002, "epoch": 3.682415976619581, "step": 3780}, {"loss": 1.384, "grad_norm": 0.6675831079483032, "learning_rate": 0.0002, "epoch": 3.6921578178275696, "step": 3790}, {"loss": 1.401, "grad_norm": 0.6480095386505127, "learning_rate": 0.0002, "epoch": 3.7018996590355577, "step": 3800}, {"loss": 1.5303, "grad_norm": 0.6559418439865112, "learning_rate": 0.0002, "epoch": 3.711641500243546, "step": 3810}, {"loss": 1.4341, "grad_norm": 0.6596545577049255, "learning_rate": 0.0002, "epoch": 3.7213833414515345, "step": 3820}, {"loss": 1.4508, "grad_norm": 0.7172950506210327, "learning_rate": 0.0002, "epoch": 3.7311251826595226, "step": 3830}, {"loss": 1.446, "grad_norm": 0.796148419380188, "learning_rate": 0.0002, "epoch": 3.740867023867511, "step": 3840}, {"loss": 1.4992, "grad_norm": 0.6600322723388672, "learning_rate": 0.0002, "epoch": 3.7506088650754994, "step": 3850}, {"loss": 1.4201, "grad_norm": 0.6776387691497803, "learning_rate": 0.0002, "epoch": 3.7603507062834876, "step": 3860}, {"loss": 1.3893, "grad_norm": 0.7768304347991943, "learning_rate": 0.0002, "epoch": 3.770092547491476, "step": 3870}, {"loss": 1.4886, "grad_norm": 1.0579794645309448, "learning_rate": 0.0002, "epoch": 3.7798343886994643, "step": 3880}, {"loss": 1.4556, "grad_norm": 0.6757252812385559, "learning_rate": 0.0002, "epoch": 3.7895762299074525, "step": 3890}, {"loss": 1.4647, "grad_norm": 0.6706996560096741, "learning_rate": 0.0002, "epoch": 3.799318071115441, "step": 3900}, {"loss": 1.4104, "grad_norm": 0.7026948928833008, "learning_rate": 0.0002, "epoch": 3.809059912323429, "step": 3910}, {"loss": 1.5487, "grad_norm": 0.6437768340110779, "learning_rate": 0.0002, "epoch": 3.8188017535314174, "step": 3920}, {"loss": 1.4678, "grad_norm": 0.7015706300735474, "learning_rate": 0.0002, "epoch": 3.828543594739406, "step": 3930}, {"loss": 1.4891, "grad_norm": 0.7049482464790344, "learning_rate": 0.0002, "epoch": 3.838285435947394, "step": 3940}, {"loss": 1.4208, "grad_norm": 0.6533724665641785, "learning_rate": 0.0002, "epoch": 3.8480272771553823, "step": 3950}, {"loss": 1.4435, "grad_norm": 0.7312499284744263, "learning_rate": 0.0002, "epoch": 3.857769118363371, "step": 3960}, {"loss": 1.3886, "grad_norm": 0.6858801245689392, "learning_rate": 0.0002, "epoch": 3.867510959571359, "step": 3970}, {"loss": 1.4423, "grad_norm": 0.770423173904419, "learning_rate": 0.0002, "epoch": 3.877252800779347, "step": 3980}, {"loss": 1.5029, "grad_norm": 0.6987539529800415, "learning_rate": 0.0002, "epoch": 3.886994641987336, "step": 3990}, {"loss": 1.4791, "grad_norm": 0.7072722315788269, "learning_rate": 0.0002, "epoch": 3.896736483195324, "step": 4000}, {"loss": 1.528, "grad_norm": 0.6492931842803955, "learning_rate": 0.0002, "epoch": 3.906478324403312, "step": 4010}, {"loss": 1.3824, "grad_norm": 0.7716232538223267, "learning_rate": 0.0002, "epoch": 3.9162201656113007, "step": 4020}, {"loss": 1.4758, "grad_norm": 0.722949743270874, "learning_rate": 0.0002, "epoch": 3.925962006819289, "step": 4030}, {"loss": 1.3914, "grad_norm": 0.7434365749359131, "learning_rate": 0.0002, "epoch": 3.935703848027277, "step": 4040}, {"loss": 1.4763, "grad_norm": 0.6691509485244751, "learning_rate": 0.0002, "epoch": 3.9454456892352656, "step": 4050}, {"loss": 1.4555, "grad_norm": 0.6850284337997437, "learning_rate": 0.0002, "epoch": 3.9551875304432538, "step": 4060}, {"loss": 1.5275, "grad_norm": 0.6954452991485596, "learning_rate": 0.0002, "epoch": 3.964929371651242, "step": 4070}, {"loss": 1.417, "grad_norm": 0.9316364526748657, "learning_rate": 0.0002, "epoch": 3.9746712128592305, "step": 4080}, {"loss": 1.4532, "grad_norm": 0.6908289194107056, "learning_rate": 0.0002, "epoch": 3.9844130540672187, "step": 4090}, {"loss": 1.4404, "grad_norm": 0.666782021522522, "learning_rate": 0.0002, "epoch": 3.994154895275207, "step": 4100}]}
+{"epoch": 4.9995129079396, "step": 5132, "epoch_duration": 1650.0333547592163, "total_accumulated_duration": 9224.923346281052, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}, {"eval_loss": 1.8463934659957886, "eval_runtime": 56.2401, "eval_samples_per_second": 9.015, "eval_steps_per_second": 1.138, "epoch": 2.9995129079396006, "step": 3079}, {"loss": 1.5744, "grad_norm": 0.49992576241493225, "learning_rate": 0.0002, "epoch": 3.0004870920603994, "step": 3080}, {"loss": 1.4125, "grad_norm": 0.8242783546447754, "learning_rate": 0.0002, "epoch": 3.0102289332683876, "step": 3090}, {"loss": 1.394, "grad_norm": 0.6330569386482239, "learning_rate": 0.0002, "epoch": 3.019970774476376, "step": 3100}, {"loss": 1.4942, "grad_norm": 0.566097617149353, "learning_rate": 0.0002, "epoch": 3.0297126156843643, "step": 3110}, {"loss": 1.4365, "grad_norm": 0.6337586045265198, "learning_rate": 0.0002, "epoch": 3.0394544568923525, "step": 3120}, {"loss": 1.3916, "grad_norm": 0.7339403033256531, "learning_rate": 0.0002, "epoch": 3.049196298100341, "step": 3130}, {"loss": 1.4617, "grad_norm": 0.7187346816062927, "learning_rate": 0.0002, "epoch": 3.0589381393083293, "step": 3140}, {"loss": 1.3453, "grad_norm": 0.7116255760192871, "learning_rate": 0.0002, "epoch": 3.0686799805163174, "step": 3150}, {"loss": 1.4452, "grad_norm": 0.6493807435035706, "learning_rate": 0.0002, "epoch": 3.078421821724306, "step": 3160}, {"loss": 1.351, "grad_norm": 0.6777266263961792, "learning_rate": 0.0002, "epoch": 3.088163662932294, "step": 3170}, {"loss": 1.4362, "grad_norm": 0.6342006325721741, "learning_rate": 0.0002, "epoch": 3.0979055041402823, "step": 3180}, {"loss": 1.4748, "grad_norm": 0.6608964204788208, "learning_rate": 0.0002, "epoch": 3.107647345348271, "step": 3190}, {"loss": 1.375, "grad_norm": 0.7230247259140015, "learning_rate": 0.0002, "epoch": 3.117389186556259, "step": 3200}, {"loss": 1.4049, "grad_norm": 0.650368332862854, "learning_rate": 0.0002, "epoch": 3.1271310277642472, "step": 3210}, {"loss": 1.409, "grad_norm": 0.7319342494010925, "learning_rate": 0.0002, "epoch": 3.136872868972236, "step": 3220}, {"loss": 1.3872, "grad_norm": 0.7159963846206665, "learning_rate": 0.0002, "epoch": 3.146614710180224, "step": 3230}, {"loss": 1.5076, "grad_norm": 0.8905230164527893, "learning_rate": 0.0002, "epoch": 3.156356551388212, "step": 3240}, {"loss": 1.3161, "grad_norm": 0.6920804381370544, "learning_rate": 0.0002, "epoch": 3.1660983925962007, "step": 3250}, {"loss": 1.3786, "grad_norm": 0.6782063841819763, "learning_rate": 0.0002, "epoch": 3.175840233804189, "step": 3260}, {"loss": 1.5153, "grad_norm": 0.735325276851654, "learning_rate": 0.0002, "epoch": 3.1855820750121775, "step": 3270}, {"loss": 1.4027, "grad_norm": 0.6657978296279907, "learning_rate": 0.0002, "epoch": 3.1953239162201656, "step": 3280}, {"loss": 1.3456, "grad_norm": 0.771315336227417, "learning_rate": 0.0002, "epoch": 3.205065757428154, "step": 3290}, {"loss": 1.3236, "grad_norm": 0.6492983102798462, "learning_rate": 0.0002, "epoch": 3.2148075986361424, "step": 3300}, {"loss": 1.4125, "grad_norm": 0.7513770461082458, "learning_rate": 0.0002, "epoch": 3.2245494398441306, "step": 3310}, {"loss": 1.4032, "grad_norm": 0.7091423869132996, "learning_rate": 0.0002, "epoch": 3.2342912810521187, "step": 3320}, {"loss": 1.4585, "grad_norm": 0.6663975119590759, "learning_rate": 0.0002, "epoch": 3.2440331222601073, "step": 3330}, {"loss": 1.3968, "grad_norm": 0.6813122034072876, "learning_rate": 0.0002, "epoch": 3.2537749634680955, "step": 3340}, {"loss": 1.3681, "grad_norm": 0.6602569818496704, "learning_rate": 0.0002, "epoch": 3.2635168046760836, "step": 3350}, {"loss": 1.4533, "grad_norm": 0.718270480632782, "learning_rate": 0.0002, "epoch": 3.2732586458840722, "step": 3360}, {"loss": 1.4076, "grad_norm": 0.6884173154830933, "learning_rate": 0.0002, "epoch": 3.2830004870920604, "step": 3370}, {"loss": 1.4144, "grad_norm": 0.7039775848388672, "learning_rate": 0.0002, "epoch": 3.2927423283000485, "step": 3380}, {"loss": 1.5077, "grad_norm": 0.7444299459457397, "learning_rate": 0.0002, "epoch": 3.302484169508037, "step": 3390}, {"loss": 1.4255, "grad_norm": 0.7187064290046692, "learning_rate": 0.0002, "epoch": 3.3122260107160253, "step": 3400}, {"loss": 1.3684, "grad_norm": 0.599396288394928, "learning_rate": 0.0002, "epoch": 3.3219678519240134, "step": 3410}, {"loss": 1.4819, "grad_norm": 0.7670390009880066, "learning_rate": 0.0002, "epoch": 3.331709693132002, "step": 3420}, {"loss": 1.4411, "grad_norm": 0.6654478311538696, "learning_rate": 0.0002, "epoch": 3.34145153433999, "step": 3430}, {"loss": 1.4257, "grad_norm": 0.6644385457038879, "learning_rate": 0.0002, "epoch": 3.351193375547979, "step": 3440}, {"loss": 1.4508, "grad_norm": 0.6974098086357117, "learning_rate": 0.0002, "epoch": 3.360935216755967, "step": 3450}, {"loss": 1.3807, "grad_norm": 0.7350399494171143, "learning_rate": 0.0002, "epoch": 3.370677057963955, "step": 3460}, {"loss": 1.4176, "grad_norm": 0.714721143245697, "learning_rate": 0.0002, "epoch": 3.3804188991719437, "step": 3470}, {"loss": 1.4325, "grad_norm": 0.7006027698516846, "learning_rate": 0.0002, "epoch": 3.390160740379932, "step": 3480}, {"loss": 1.4888, "grad_norm": 0.6767925024032593, "learning_rate": 0.0002, "epoch": 3.39990258158792, "step": 3490}, {"loss": 1.4116, "grad_norm": 0.6721355319023132, "learning_rate": 0.0002, "epoch": 3.4096444227959086, "step": 3500}, {"loss": 1.443, "grad_norm": 0.6845725178718567, "learning_rate": 0.0002, "epoch": 3.419386264003897, "step": 3510}, {"loss": 1.4832, "grad_norm": 0.6882196664810181, "learning_rate": 0.0002, "epoch": 3.429128105211885, "step": 3520}, {"loss": 1.4962, "grad_norm": 0.7663240432739258, "learning_rate": 0.0002, "epoch": 3.4388699464198735, "step": 3530}, {"loss": 1.4644, "grad_norm": 0.6304219365119934, "learning_rate": 0.0002, "epoch": 3.4486117876278617, "step": 3540}, {"loss": 1.4918, "grad_norm": 0.668678879737854, "learning_rate": 0.0002, "epoch": 3.45835362883585, "step": 3550}, {"loss": 1.4874, "grad_norm": 0.7526912093162537, "learning_rate": 0.0002, "epoch": 3.4680954700438384, "step": 3560}, {"loss": 1.4249, "grad_norm": 1.089495301246643, "learning_rate": 0.0002, "epoch": 3.4778373112518266, "step": 3570}, {"loss": 1.3871, "grad_norm": 0.7282902002334595, "learning_rate": 0.0002, "epoch": 3.4875791524598148, "step": 3580}, {"loss": 1.5077, "grad_norm": 0.6540156602859497, "learning_rate": 0.0002, "epoch": 3.4973209936678034, "step": 3590}, {"loss": 1.4367, "grad_norm": 0.6449568867683411, "learning_rate": 0.0002, "epoch": 3.5070628348757915, "step": 3600}, {"loss": 1.4532, "grad_norm": 0.7262216210365295, "learning_rate": 0.0002, "epoch": 3.5168046760837797, "step": 3610}, {"loss": 1.4374, "grad_norm": 0.6048615574836731, "learning_rate": 0.0002, "epoch": 3.5265465172917683, "step": 3620}, {"loss": 1.3877, "grad_norm": 0.6780537366867065, "learning_rate": 0.0002, "epoch": 3.5362883584997564, "step": 3630}, {"loss": 1.422, "grad_norm": 0.6851925253868103, "learning_rate": 0.0002, "epoch": 3.5460301997077446, "step": 3640}, {"loss": 1.3425, "grad_norm": 0.6530634164810181, "learning_rate": 0.0002, "epoch": 3.555772040915733, "step": 3650}, {"loss": 1.4879, "grad_norm": 0.7193992733955383, "learning_rate": 0.0002, "epoch": 3.5655138821237213, "step": 3660}, {"loss": 1.4555, "grad_norm": 0.767496645450592, "learning_rate": 0.0002, "epoch": 3.5752557233317095, "step": 3670}, {"loss": 1.4824, "grad_norm": 0.6912919282913208, "learning_rate": 0.0002, "epoch": 3.584997564539698, "step": 3680}, {"loss": 1.4497, "grad_norm": 0.7383436560630798, "learning_rate": 0.0002, "epoch": 3.5947394057476862, "step": 3690}, {"loss": 1.4822, "grad_norm": 0.6746662855148315, "learning_rate": 0.0002, "epoch": 3.6044812469556744, "step": 3700}, {"loss": 1.4904, "grad_norm": 0.6885138750076294, "learning_rate": 0.0002, "epoch": 3.614223088163663, "step": 3710}, {"loss": 1.4044, "grad_norm": 0.6694392561912537, "learning_rate": 0.0002, "epoch": 3.623964929371651, "step": 3720}, {"loss": 1.3719, "grad_norm": 0.812358021736145, "learning_rate": 0.0002, "epoch": 3.6337067705796393, "step": 3730}, {"loss": 1.4603, "grad_norm": 0.7267130017280579, "learning_rate": 0.0002, "epoch": 3.643448611787628, "step": 3740}, {"loss": 1.4574, "grad_norm": 0.6958749294281006, "learning_rate": 0.0002, "epoch": 3.653190452995616, "step": 3750}, {"loss": 1.4346, "grad_norm": 0.6805673241615295, "learning_rate": 0.0002, "epoch": 3.6629322942036042, "step": 3760}, {"loss": 1.4338, "grad_norm": 0.7184410095214844, "learning_rate": 0.0002, "epoch": 3.672674135411593, "step": 3770}, {"loss": 1.3935, "grad_norm": 0.7716330289840698, "learning_rate": 0.0002, "epoch": 3.682415976619581, "step": 3780}, {"loss": 1.384, "grad_norm": 0.6675831079483032, "learning_rate": 0.0002, "epoch": 3.6921578178275696, "step": 3790}, {"loss": 1.401, "grad_norm": 0.6480095386505127, "learning_rate": 0.0002, "epoch": 3.7018996590355577, "step": 3800}, {"loss": 1.5303, "grad_norm": 0.6559418439865112, "learning_rate": 0.0002, "epoch": 3.711641500243546, "step": 3810}, {"loss": 1.4341, "grad_norm": 0.6596545577049255, "learning_rate": 0.0002, "epoch": 3.7213833414515345, "step": 3820}, {"loss": 1.4508, "grad_norm": 0.7172950506210327, "learning_rate": 0.0002, "epoch": 3.7311251826595226, "step": 3830}, {"loss": 1.446, "grad_norm": 0.796148419380188, "learning_rate": 0.0002, "epoch": 3.740867023867511, "step": 3840}, {"loss": 1.4992, "grad_norm": 0.6600322723388672, "learning_rate": 0.0002, "epoch": 3.7506088650754994, "step": 3850}, {"loss": 1.4201, "grad_norm": 0.6776387691497803, "learning_rate": 0.0002, "epoch": 3.7603507062834876, "step": 3860}, {"loss": 1.3893, "grad_norm": 0.7768304347991943, "learning_rate": 0.0002, "epoch": 3.770092547491476, "step": 3870}, {"loss": 1.4886, "grad_norm": 1.0579794645309448, "learning_rate": 0.0002, "epoch": 3.7798343886994643, "step": 3880}, {"loss": 1.4556, "grad_norm": 0.6757252812385559, "learning_rate": 0.0002, "epoch": 3.7895762299074525, "step": 3890}, {"loss": 1.4647, "grad_norm": 0.6706996560096741, "learning_rate": 0.0002, "epoch": 3.799318071115441, "step": 3900}, {"loss": 1.4104, "grad_norm": 0.7026948928833008, "learning_rate": 0.0002, "epoch": 3.809059912323429, "step": 3910}, {"loss": 1.5487, "grad_norm": 0.6437768340110779, "learning_rate": 0.0002, "epoch": 3.8188017535314174, "step": 3920}, {"loss": 1.4678, "grad_norm": 0.7015706300735474, "learning_rate": 0.0002, "epoch": 3.828543594739406, "step": 3930}, {"loss": 1.4891, "grad_norm": 0.7049482464790344, "learning_rate": 0.0002, "epoch": 3.838285435947394, "step": 3940}, {"loss": 1.4208, "grad_norm": 0.6533724665641785, "learning_rate": 0.0002, "epoch": 3.8480272771553823, "step": 3950}, {"loss": 1.4435, "grad_norm": 0.7312499284744263, "learning_rate": 0.0002, "epoch": 3.857769118363371, "step": 3960}, {"loss": 1.3886, "grad_norm": 0.6858801245689392, "learning_rate": 0.0002, "epoch": 3.867510959571359, "step": 3970}, {"loss": 1.4423, "grad_norm": 0.770423173904419, "learning_rate": 0.0002, "epoch": 3.877252800779347, "step": 3980}, {"loss": 1.5029, "grad_norm": 0.6987539529800415, "learning_rate": 0.0002, "epoch": 3.886994641987336, "step": 3990}, {"loss": 1.4791, "grad_norm": 0.7072722315788269, "learning_rate": 0.0002, "epoch": 3.896736483195324, "step": 4000}, {"loss": 1.528, "grad_norm": 0.6492931842803955, "learning_rate": 0.0002, "epoch": 3.906478324403312, "step": 4010}, {"loss": 1.3824, "grad_norm": 0.7716232538223267, "learning_rate": 0.0002, "epoch": 3.9162201656113007, "step": 4020}, {"loss": 1.4758, "grad_norm": 0.722949743270874, "learning_rate": 0.0002, "epoch": 3.925962006819289, "step": 4030}, {"loss": 1.3914, "grad_norm": 0.7434365749359131, "learning_rate": 0.0002, "epoch": 3.935703848027277, "step": 4040}, {"loss": 1.4763, "grad_norm": 0.6691509485244751, "learning_rate": 0.0002, "epoch": 3.9454456892352656, "step": 4050}, {"loss": 1.4555, "grad_norm": 0.6850284337997437, "learning_rate": 0.0002, "epoch": 3.9551875304432538, "step": 4060}, {"loss": 1.5275, "grad_norm": 0.6954452991485596, "learning_rate": 0.0002, "epoch": 3.964929371651242, "step": 4070}, {"loss": 1.417, "grad_norm": 0.9316364526748657, "learning_rate": 0.0002, "epoch": 3.9746712128592305, "step": 4080}, {"loss": 1.4532, "grad_norm": 0.6908289194107056, "learning_rate": 0.0002, "epoch": 3.9844130540672187, "step": 4090}, {"loss": 1.4404, "grad_norm": 0.666782021522522, "learning_rate": 0.0002, "epoch": 3.994154895275207, "step": 4100}, {"eval_loss": 1.9233275651931763, "eval_runtime": 55.9536, "eval_samples_per_second": 9.061, "eval_steps_per_second": 1.144, "epoch": 4.0, "step": 4106}, {"loss": 1.3489, "grad_norm": 0.7726166248321533, "learning_rate": 0.0002, "epoch": 4.003896736483195, "step": 4110}, {"loss": 1.1415, "grad_norm": 1.1338967084884644, "learning_rate": 0.0002, "epoch": 4.013638577691184, "step": 4120}, {"loss": 1.2212, "grad_norm": 0.9530029296875, "learning_rate": 0.0002, "epoch": 4.023380418899172, "step": 4130}, {"loss": 1.2002, "grad_norm": 1.1058554649353027, "learning_rate": 0.0002, "epoch": 4.03312226010716, "step": 4140}, {"loss": 1.2381, "grad_norm": 0.8765049576759338, "learning_rate": 0.0002, "epoch": 4.042864101315149, "step": 4150}, {"loss": 1.2708, "grad_norm": 1.1774667501449585, "learning_rate": 0.0002, "epoch": 4.052605942523137, "step": 4160}, {"loss": 1.2116, "grad_norm": 0.9301433563232422, "learning_rate": 0.0002, "epoch": 4.062347783731125, "step": 4170}, {"loss": 1.1807, "grad_norm": 1.0196778774261475, "learning_rate": 0.0002, "epoch": 4.072089624939114, "step": 4180}, {"loss": 1.2602, "grad_norm": 1.1380577087402344, "learning_rate": 0.0002, "epoch": 4.081831466147102, "step": 4190}, {"loss": 1.2521, "grad_norm": 0.9121319651603699, "learning_rate": 0.0002, "epoch": 4.09157330735509, "step": 4200}, {"loss": 1.1747, "grad_norm": 0.9495378732681274, "learning_rate": 0.0002, "epoch": 4.101315148563079, "step": 4210}, {"loss": 1.1829, "grad_norm": 0.8058680295944214, "learning_rate": 0.0002, "epoch": 4.1110569897710665, "step": 4220}, {"loss": 1.1732, "grad_norm": 1.000887393951416, "learning_rate": 0.0002, "epoch": 4.120798830979055, "step": 4230}, {"loss": 1.1947, "grad_norm": 0.9529102444648743, "learning_rate": 0.0002, "epoch": 4.130540672187044, "step": 4240}, {"loss": 1.2104, "grad_norm": 1.0257115364074707, "learning_rate": 0.0002, "epoch": 4.140282513395031, "step": 4250}, {"loss": 1.2293, "grad_norm": 0.9590303897857666, "learning_rate": 0.0002, "epoch": 4.15002435460302, "step": 4260}, {"loss": 1.1918, "grad_norm": 1.065291166305542, "learning_rate": 0.0002, "epoch": 4.159766195811009, "step": 4270}, {"loss": 1.2323, "grad_norm": 0.8819697499275208, "learning_rate": 0.0002, "epoch": 4.169508037018996, "step": 4280}, {"loss": 1.2167, "grad_norm": 1.0335261821746826, "learning_rate": 0.0002, "epoch": 4.179249878226985, "step": 4290}, {"loss": 1.2131, "grad_norm": 0.8872809410095215, "learning_rate": 0.0002, "epoch": 4.1889917194349735, "step": 4300}, {"loss": 1.2794, "grad_norm": 0.9883159399032593, "learning_rate": 0.0002, "epoch": 4.198733560642961, "step": 4310}, {"loss": 1.2544, "grad_norm": 1.0254192352294922, "learning_rate": 0.0002, "epoch": 4.20847540185095, "step": 4320}, {"loss": 1.2595, "grad_norm": 0.9432600736618042, "learning_rate": 0.0002, "epoch": 4.218217243058938, "step": 4330}, {"loss": 1.2684, "grad_norm": 1.1008676290512085, "learning_rate": 0.0002, "epoch": 4.227959084266926, "step": 4340}, {"loss": 1.2149, "grad_norm": 1.0829699039459229, "learning_rate": 0.0002, "epoch": 4.237700925474915, "step": 4350}, {"loss": 1.2621, "grad_norm": 1.016847848892212, "learning_rate": 0.0002, "epoch": 4.247442766682903, "step": 4360}, {"loss": 1.2375, "grad_norm": 0.8924864530563354, "learning_rate": 0.0002, "epoch": 4.257184607890891, "step": 4370}, {"loss": 1.1987, "grad_norm": 0.9300530552864075, "learning_rate": 0.0002, "epoch": 4.26692644909888, "step": 4380}, {"loss": 1.1696, "grad_norm": 0.9684814810752869, "learning_rate": 0.0002, "epoch": 4.276668290306868, "step": 4390}, {"loss": 1.2006, "grad_norm": 0.9916250705718994, "learning_rate": 0.0002, "epoch": 4.286410131514856, "step": 4400}, {"loss": 1.2402, "grad_norm": 0.903680145740509, "learning_rate": 0.0002, "epoch": 4.2961519727228445, "step": 4410}, {"loss": 1.2022, "grad_norm": 0.8713505268096924, "learning_rate": 0.0002, "epoch": 4.305893813930833, "step": 4420}, {"loss": 1.1957, "grad_norm": 0.9983905553817749, "learning_rate": 0.0002, "epoch": 4.315635655138821, "step": 4430}, {"loss": 1.2676, "grad_norm": 1.1689040660858154, "learning_rate": 0.0002, "epoch": 4.3253774963468095, "step": 4440}, {"loss": 1.2166, "grad_norm": 0.9316853880882263, "learning_rate": 0.0002, "epoch": 4.335119337554798, "step": 4450}, {"loss": 1.222, "grad_norm": 0.9175887107849121, "learning_rate": 0.0002, "epoch": 4.344861178762786, "step": 4460}, {"loss": 1.2571, "grad_norm": 0.9348906874656677, "learning_rate": 0.0002, "epoch": 4.354603019970774, "step": 4470}, {"loss": 1.2764, "grad_norm": 0.9727016687393188, "learning_rate": 0.0002, "epoch": 4.364344861178763, "step": 4480}, {"loss": 1.2616, "grad_norm": 0.9843429923057556, "learning_rate": 0.0002, "epoch": 4.374086702386751, "step": 4490}, {"loss": 1.2488, "grad_norm": 0.9615852236747742, "learning_rate": 0.0002, "epoch": 4.383828543594739, "step": 4500}, {"loss": 1.1718, "grad_norm": 0.9688583612442017, "learning_rate": 0.0002, "epoch": 4.393570384802728, "step": 4510}, {"loss": 1.2546, "grad_norm": 0.9933668375015259, "learning_rate": 0.0002, "epoch": 4.403312226010716, "step": 4520}, {"loss": 1.2355, "grad_norm": 1.0626686811447144, "learning_rate": 0.0002, "epoch": 4.413054067218704, "step": 4530}, {"loss": 1.2425, "grad_norm": 0.9536267518997192, "learning_rate": 0.0002, "epoch": 4.422795908426693, "step": 4540}, {"loss": 1.2562, "grad_norm": 0.9777140021324158, "learning_rate": 0.0002, "epoch": 4.432537749634681, "step": 4550}, {"loss": 1.2878, "grad_norm": 0.980780839920044, "learning_rate": 0.0002, "epoch": 4.442279590842669, "step": 4560}, {"loss": 1.2597, "grad_norm": 1.0147196054458618, "learning_rate": 0.0002, "epoch": 4.452021432050658, "step": 4570}, {"loss": 1.2148, "grad_norm": 0.9763361811637878, "learning_rate": 0.0002, "epoch": 4.461763273258645, "step": 4580}, {"loss": 1.3076, "grad_norm": 1.0300798416137695, "learning_rate": 0.0002, "epoch": 4.471505114466634, "step": 4590}, {"loss": 1.2665, "grad_norm": 0.8833121657371521, "learning_rate": 0.0002, "epoch": 4.481246955674623, "step": 4600}, {"loss": 1.1899, "grad_norm": 1.1214020252227783, "learning_rate": 0.0002, "epoch": 4.490988796882611, "step": 4610}, {"loss": 1.2579, "grad_norm": 0.8843787908554077, "learning_rate": 0.0002, "epoch": 4.500730638090599, "step": 4620}, {"loss": 1.2633, "grad_norm": 0.9942020773887634, "learning_rate": 0.0002, "epoch": 4.5104724792985875, "step": 4630}, {"loss": 1.3172, "grad_norm": 1.0033202171325684, "learning_rate": 0.0002, "epoch": 4.520214320506576, "step": 4640}, {"loss": 1.2024, "grad_norm": 0.8767235279083252, "learning_rate": 0.0002, "epoch": 4.529956161714564, "step": 4650}, {"loss": 1.2714, "grad_norm": 1.0117276906967163, "learning_rate": 0.0002, "epoch": 4.539698002922552, "step": 4660}, {"loss": 1.2911, "grad_norm": 1.2787362337112427, "learning_rate": 0.0002, "epoch": 4.549439844130541, "step": 4670}, {"loss": 1.2603, "grad_norm": 0.8824878931045532, "learning_rate": 0.0002, "epoch": 4.559181685338529, "step": 4680}, {"loss": 1.2905, "grad_norm": 0.9209560751914978, "learning_rate": 0.0002, "epoch": 4.568923526546517, "step": 4690}, {"loss": 1.1916, "grad_norm": 1.1064010858535767, "learning_rate": 0.0002, "epoch": 4.578665367754506, "step": 4700}, {"loss": 1.2217, "grad_norm": 0.8914572596549988, "learning_rate": 0.0002, "epoch": 4.588407208962494, "step": 4710}, {"loss": 1.2861, "grad_norm": 1.0412265062332153, "learning_rate": 0.0002, "epoch": 4.598149050170482, "step": 4720}, {"loss": 1.262, "grad_norm": 1.1950221061706543, "learning_rate": 0.0002, "epoch": 4.607890891378471, "step": 4730}, {"loss": 1.2659, "grad_norm": 0.8938062787055969, "learning_rate": 0.0002, "epoch": 4.617632732586459, "step": 4740}, {"loss": 1.2621, "grad_norm": 0.9849569201469421, "learning_rate": 0.0002, "epoch": 4.627374573794447, "step": 4750}, {"loss": 1.2341, "grad_norm": 1.0081515312194824, "learning_rate": 0.0002, "epoch": 4.637116415002436, "step": 4760}, {"loss": 1.2023, "grad_norm": 0.8566309213638306, "learning_rate": 0.0002, "epoch": 4.6468582562104235, "step": 4770}, {"loss": 1.2723, "grad_norm": 1.1750118732452393, "learning_rate": 0.0002, "epoch": 4.656600097418412, "step": 4780}, {"loss": 1.2537, "grad_norm": 0.925502598285675, "learning_rate": 0.0002, "epoch": 4.666341938626401, "step": 4790}, {"loss": 1.2146, "grad_norm": 1.0402472019195557, "learning_rate": 0.0002, "epoch": 4.676083779834388, "step": 4800}, {"loss": 1.2555, "grad_norm": 0.9772472977638245, "learning_rate": 0.0002, "epoch": 4.685825621042377, "step": 4810}, {"loss": 1.2667, "grad_norm": 0.9082779288291931, "learning_rate": 0.0002, "epoch": 4.695567462250366, "step": 4820}, {"loss": 1.2465, "grad_norm": 0.8026862740516663, "learning_rate": 0.0002, "epoch": 4.705309303458353, "step": 4830}, {"loss": 1.3369, "grad_norm": 1.1631089448928833, "learning_rate": 0.0002, "epoch": 4.715051144666342, "step": 4840}, {"loss": 1.261, "grad_norm": 0.9384787678718567, "learning_rate": 0.0002, "epoch": 4.7247929858743305, "step": 4850}, {"loss": 1.2588, "grad_norm": 1.2151581048965454, "learning_rate": 0.0002, "epoch": 4.734534827082318, "step": 4860}, {"loss": 1.363, "grad_norm": 0.9679436087608337, "learning_rate": 0.0002, "epoch": 4.744276668290307, "step": 4870}, {"loss": 1.3292, "grad_norm": 0.8352158069610596, "learning_rate": 0.0002, "epoch": 4.754018509498295, "step": 4880}, {"loss": 1.3056, "grad_norm": 1.0205804109573364, "learning_rate": 0.0002, "epoch": 4.763760350706283, "step": 4890}, {"loss": 1.223, "grad_norm": 0.9814772605895996, "learning_rate": 0.0002, "epoch": 4.773502191914272, "step": 4900}, {"loss": 1.3114, "grad_norm": 1.002854347229004, "learning_rate": 0.0002, "epoch": 4.78324403312226, "step": 4910}, {"loss": 1.3143, "grad_norm": 1.1609505414962769, "learning_rate": 0.0002, "epoch": 4.792985874330248, "step": 4920}, {"loss": 1.3166, "grad_norm": 0.9354982376098633, "learning_rate": 0.0002, "epoch": 4.802727715538237, "step": 4930}, {"loss": 1.2978, "grad_norm": 0.9761685729026794, "learning_rate": 0.0002, "epoch": 4.812469556746225, "step": 4940}, {"loss": 1.2709, "grad_norm": 1.0604596138000488, "learning_rate": 0.0002, "epoch": 4.822211397954213, "step": 4950}, {"loss": 1.2765, "grad_norm": 1.0902808904647827, "learning_rate": 0.0002, "epoch": 4.8319532391622015, "step": 4960}, {"loss": 1.3073, "grad_norm": 1.0174955129623413, "learning_rate": 0.0002, "epoch": 4.84169508037019, "step": 4970}, {"loss": 1.3141, "grad_norm": 1.0995253324508667, "learning_rate": 0.0002, "epoch": 4.851436921578179, "step": 4980}, {"loss": 1.3006, "grad_norm": 0.880993127822876, "learning_rate": 0.0002, "epoch": 4.8611787627861665, "step": 4990}, {"loss": 1.2547, "grad_norm": 0.9472237825393677, "learning_rate": 0.0002, "epoch": 4.870920603994155, "step": 5000}, {"loss": 1.4078, "grad_norm": 0.9504236578941345, "learning_rate": 0.0002, "epoch": 4.880662445202143, "step": 5010}, {"loss": 1.2791, "grad_norm": 1.1261742115020752, "learning_rate": 0.0002, "epoch": 4.890404286410131, "step": 5020}, {"loss": 1.3707, "grad_norm": 0.904674768447876, "learning_rate": 0.0002, "epoch": 4.90014612761812, "step": 5030}, {"loss": 1.2762, "grad_norm": 0.8828991055488586, "learning_rate": 0.0002, "epoch": 4.909887968826109, "step": 5040}, {"loss": 1.2905, "grad_norm": 1.0156532526016235, "learning_rate": 0.0002, "epoch": 4.919629810034096, "step": 5050}, {"loss": 1.3079, "grad_norm": 0.8975168466567993, "learning_rate": 0.0002, "epoch": 4.929371651242085, "step": 5060}, {"loss": 1.3322, "grad_norm": 0.9787213802337646, "learning_rate": 0.0002, "epoch": 4.939113492450073, "step": 5070}, {"loss": 1.2533, "grad_norm": 1.0801568031311035, "learning_rate": 0.0002, "epoch": 4.948855333658061, "step": 5080}, {"loss": 1.238, "grad_norm": 1.0655089616775513, "learning_rate": 0.0002, "epoch": 4.95859717486605, "step": 5090}, {"loss": 1.2449, "grad_norm": 0.8941320180892944, "learning_rate": 0.0002, "epoch": 4.968339016074038, "step": 5100}, {"loss": 1.2846, "grad_norm": 1.050621747970581, "learning_rate": 0.0002, "epoch": 4.978080857282026, "step": 5110}, {"loss": 1.3791, "grad_norm": 0.9724781513214111, "learning_rate": 0.0002, "epoch": 4.987822698490015, "step": 5120}, {"loss": 1.292, "grad_norm": 0.9850538969039917, "learning_rate": 0.0002, "epoch": 4.997564539698003, "step": 5130}]}
+{"epoch": 6.0, "step": 6159, "epoch_duration": 1656.5674257278442, "total_accumulated_duration": 10881.490772008896, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}, {"eval_loss": 1.8463934659957886, "eval_runtime": 56.2401, "eval_samples_per_second": 9.015, "eval_steps_per_second": 1.138, "epoch": 2.9995129079396006, "step": 3079}, {"loss": 1.5744, "grad_norm": 0.49992576241493225, "learning_rate": 0.0002, "epoch": 3.0004870920603994, "step": 3080}, {"loss": 1.4125, "grad_norm": 0.8242783546447754, "learning_rate": 0.0002, "epoch": 3.0102289332683876, "step": 3090}, {"loss": 1.394, "grad_norm": 0.6330569386482239, "learning_rate": 0.0002, "epoch": 3.019970774476376, "step": 3100}, {"loss": 1.4942, "grad_norm": 0.566097617149353, "learning_rate": 0.0002, "epoch": 3.0297126156843643, "step": 3110}, {"loss": 1.4365, "grad_norm": 0.6337586045265198, "learning_rate": 0.0002, "epoch": 3.0394544568923525, "step": 3120}, {"loss": 1.3916, "grad_norm": 0.7339403033256531, "learning_rate": 0.0002, "epoch": 3.049196298100341, "step": 3130}, {"loss": 1.4617, "grad_norm": 0.7187346816062927, "learning_rate": 0.0002, "epoch": 3.0589381393083293, "step": 3140}, {"loss": 1.3453, "grad_norm": 0.7116255760192871, "learning_rate": 0.0002, "epoch": 3.0686799805163174, "step": 3150}, {"loss": 1.4452, "grad_norm": 0.6493807435035706, "learning_rate": 0.0002, "epoch": 3.078421821724306, "step": 3160}, {"loss": 1.351, "grad_norm": 0.6777266263961792, "learning_rate": 0.0002, "epoch": 3.088163662932294, "step": 3170}, {"loss": 1.4362, "grad_norm": 0.6342006325721741, "learning_rate": 0.0002, "epoch": 3.0979055041402823, "step": 3180}, {"loss": 1.4748, "grad_norm": 0.6608964204788208, "learning_rate": 0.0002, "epoch": 3.107647345348271, "step": 3190}, {"loss": 1.375, "grad_norm": 0.7230247259140015, "learning_rate": 0.0002, "epoch": 3.117389186556259, "step": 3200}, {"loss": 1.4049, "grad_norm": 0.650368332862854, "learning_rate": 0.0002, "epoch": 3.1271310277642472, "step": 3210}, {"loss": 1.409, "grad_norm": 0.7319342494010925, "learning_rate": 0.0002, "epoch": 3.136872868972236, "step": 3220}, {"loss": 1.3872, "grad_norm": 0.7159963846206665, "learning_rate": 0.0002, "epoch": 3.146614710180224, "step": 3230}, {"loss": 1.5076, "grad_norm": 0.8905230164527893, "learning_rate": 0.0002, "epoch": 3.156356551388212, "step": 3240}, {"loss": 1.3161, "grad_norm": 0.6920804381370544, "learning_rate": 0.0002, "epoch": 3.1660983925962007, "step": 3250}, {"loss": 1.3786, "grad_norm": 0.6782063841819763, "learning_rate": 0.0002, "epoch": 3.175840233804189, "step": 3260}, {"loss": 1.5153, "grad_norm": 0.735325276851654, "learning_rate": 0.0002, "epoch": 3.1855820750121775, "step": 3270}, {"loss": 1.4027, "grad_norm": 0.6657978296279907, "learning_rate": 0.0002, "epoch": 3.1953239162201656, "step": 3280}, {"loss": 1.3456, "grad_norm": 0.771315336227417, "learning_rate": 0.0002, "epoch": 3.205065757428154, "step": 3290}, {"loss": 1.3236, "grad_norm": 0.6492983102798462, "learning_rate": 0.0002, "epoch": 3.2148075986361424, "step": 3300}, {"loss": 1.4125, "grad_norm": 0.7513770461082458, "learning_rate": 0.0002, "epoch": 3.2245494398441306, "step": 3310}, {"loss": 1.4032, "grad_norm": 0.7091423869132996, "learning_rate": 0.0002, "epoch": 3.2342912810521187, "step": 3320}, {"loss": 1.4585, "grad_norm": 0.6663975119590759, "learning_rate": 0.0002, "epoch": 3.2440331222601073, "step": 3330}, {"loss": 1.3968, "grad_norm": 0.6813122034072876, "learning_rate": 0.0002, "epoch": 3.2537749634680955, "step": 3340}, {"loss": 1.3681, "grad_norm": 0.6602569818496704, "learning_rate": 0.0002, "epoch": 3.2635168046760836, "step": 3350}, {"loss": 1.4533, "grad_norm": 0.718270480632782, "learning_rate": 0.0002, "epoch": 3.2732586458840722, "step": 3360}, {"loss": 1.4076, "grad_norm": 0.6884173154830933, "learning_rate": 0.0002, "epoch": 3.2830004870920604, "step": 3370}, {"loss": 1.4144, "grad_norm": 0.7039775848388672, "learning_rate": 0.0002, "epoch": 3.2927423283000485, "step": 3380}, {"loss": 1.5077, "grad_norm": 0.7444299459457397, "learning_rate": 0.0002, "epoch": 3.302484169508037, "step": 3390}, {"loss": 1.4255, "grad_norm": 0.7187064290046692, "learning_rate": 0.0002, "epoch": 3.3122260107160253, "step": 3400}, {"loss": 1.3684, "grad_norm": 0.599396288394928, "learning_rate": 0.0002, "epoch": 3.3219678519240134, "step": 3410}, {"loss": 1.4819, "grad_norm": 0.7670390009880066, "learning_rate": 0.0002, "epoch": 3.331709693132002, "step": 3420}, {"loss": 1.4411, "grad_norm": 0.6654478311538696, "learning_rate": 0.0002, "epoch": 3.34145153433999, "step": 3430}, {"loss": 1.4257, "grad_norm": 0.6644385457038879, "learning_rate": 0.0002, "epoch": 3.351193375547979, "step": 3440}, {"loss": 1.4508, "grad_norm": 0.6974098086357117, "learning_rate": 0.0002, "epoch": 3.360935216755967, "step": 3450}, {"loss": 1.3807, "grad_norm": 0.7350399494171143, "learning_rate": 0.0002, "epoch": 3.370677057963955, "step": 3460}, {"loss": 1.4176, "grad_norm": 0.714721143245697, "learning_rate": 0.0002, "epoch": 3.3804188991719437, "step": 3470}, {"loss": 1.4325, "grad_norm": 0.7006027698516846, "learning_rate": 0.0002, "epoch": 3.390160740379932, "step": 3480}, {"loss": 1.4888, "grad_norm": 0.6767925024032593, "learning_rate": 0.0002, "epoch": 3.39990258158792, "step": 3490}, {"loss": 1.4116, "grad_norm": 0.6721355319023132, "learning_rate": 0.0002, "epoch": 3.4096444227959086, "step": 3500}, {"loss": 1.443, "grad_norm": 0.6845725178718567, "learning_rate": 0.0002, "epoch": 3.419386264003897, "step": 3510}, {"loss": 1.4832, "grad_norm": 0.6882196664810181, "learning_rate": 0.0002, "epoch": 3.429128105211885, "step": 3520}, {"loss": 1.4962, "grad_norm": 0.7663240432739258, "learning_rate": 0.0002, "epoch": 3.4388699464198735, "step": 3530}, {"loss": 1.4644, "grad_norm": 0.6304219365119934, "learning_rate": 0.0002, "epoch": 3.4486117876278617, "step": 3540}, {"loss": 1.4918, "grad_norm": 0.668678879737854, "learning_rate": 0.0002, "epoch": 3.45835362883585, "step": 3550}, {"loss": 1.4874, "grad_norm": 0.7526912093162537, "learning_rate": 0.0002, "epoch": 3.4680954700438384, "step": 3560}, {"loss": 1.4249, "grad_norm": 1.089495301246643, "learning_rate": 0.0002, "epoch": 3.4778373112518266, "step": 3570}, {"loss": 1.3871, "grad_norm": 0.7282902002334595, "learning_rate": 0.0002, "epoch": 3.4875791524598148, "step": 3580}, {"loss": 1.5077, "grad_norm": 0.6540156602859497, "learning_rate": 0.0002, "epoch": 3.4973209936678034, "step": 3590}, {"loss": 1.4367, "grad_norm": 0.6449568867683411, "learning_rate": 0.0002, "epoch": 3.5070628348757915, "step": 3600}, {"loss": 1.4532, "grad_norm": 0.7262216210365295, "learning_rate": 0.0002, "epoch": 3.5168046760837797, "step": 3610}, {"loss": 1.4374, "grad_norm": 0.6048615574836731, "learning_rate": 0.0002, "epoch": 3.5265465172917683, "step": 3620}, {"loss": 1.3877, "grad_norm": 0.6780537366867065, "learning_rate": 0.0002, "epoch": 3.5362883584997564, "step": 3630}, {"loss": 1.422, "grad_norm": 0.6851925253868103, "learning_rate": 0.0002, "epoch": 3.5460301997077446, "step": 3640}, {"loss": 1.3425, "grad_norm": 0.6530634164810181, "learning_rate": 0.0002, "epoch": 3.555772040915733, "step": 3650}, {"loss": 1.4879, "grad_norm": 0.7193992733955383, "learning_rate": 0.0002, "epoch": 3.5655138821237213, "step": 3660}, {"loss": 1.4555, "grad_norm": 0.767496645450592, "learning_rate": 0.0002, "epoch": 3.5752557233317095, "step": 3670}, {"loss": 1.4824, "grad_norm": 0.6912919282913208, "learning_rate": 0.0002, "epoch": 3.584997564539698, "step": 3680}, {"loss": 1.4497, "grad_norm": 0.7383436560630798, "learning_rate": 0.0002, "epoch": 3.5947394057476862, "step": 3690}, {"loss": 1.4822, "grad_norm": 0.6746662855148315, "learning_rate": 0.0002, "epoch": 3.6044812469556744, "step": 3700}, {"loss": 1.4904, "grad_norm": 0.6885138750076294, "learning_rate": 0.0002, "epoch": 3.614223088163663, "step": 3710}, {"loss": 1.4044, "grad_norm": 0.6694392561912537, "learning_rate": 0.0002, "epoch": 3.623964929371651, "step": 3720}, {"loss": 1.3719, "grad_norm": 0.812358021736145, "learning_rate": 0.0002, "epoch": 3.6337067705796393, "step": 3730}, {"loss": 1.4603, "grad_norm": 0.7267130017280579, "learning_rate": 0.0002, "epoch": 3.643448611787628, "step": 3740}, {"loss": 1.4574, "grad_norm": 0.6958749294281006, "learning_rate": 0.0002, "epoch": 3.653190452995616, "step": 3750}, {"loss": 1.4346, "grad_norm": 0.6805673241615295, "learning_rate": 0.0002, "epoch": 3.6629322942036042, "step": 3760}, {"loss": 1.4338, "grad_norm": 0.7184410095214844, "learning_rate": 0.0002, "epoch": 3.672674135411593, "step": 3770}, {"loss": 1.3935, "grad_norm": 0.7716330289840698, "learning_rate": 0.0002, "epoch": 3.682415976619581, "step": 3780}, {"loss": 1.384, "grad_norm": 0.6675831079483032, "learning_rate": 0.0002, "epoch": 3.6921578178275696, "step": 3790}, {"loss": 1.401, "grad_norm": 0.6480095386505127, "learning_rate": 0.0002, "epoch": 3.7018996590355577, "step": 3800}, {"loss": 1.5303, "grad_norm": 0.6559418439865112, "learning_rate": 0.0002, "epoch": 3.711641500243546, "step": 3810}, {"loss": 1.4341, "grad_norm": 0.6596545577049255, "learning_rate": 0.0002, "epoch": 3.7213833414515345, "step": 3820}, {"loss": 1.4508, "grad_norm": 0.7172950506210327, "learning_rate": 0.0002, "epoch": 3.7311251826595226, "step": 3830}, {"loss": 1.446, "grad_norm": 0.796148419380188, "learning_rate": 0.0002, "epoch": 3.740867023867511, "step": 3840}, {"loss": 1.4992, "grad_norm": 0.6600322723388672, "learning_rate": 0.0002, "epoch": 3.7506088650754994, "step": 3850}, {"loss": 1.4201, "grad_norm": 0.6776387691497803, "learning_rate": 0.0002, "epoch": 3.7603507062834876, "step": 3860}, {"loss": 1.3893, "grad_norm": 0.7768304347991943, "learning_rate": 0.0002, "epoch": 3.770092547491476, "step": 3870}, {"loss": 1.4886, "grad_norm": 1.0579794645309448, "learning_rate": 0.0002, "epoch": 3.7798343886994643, "step": 3880}, {"loss": 1.4556, "grad_norm": 0.6757252812385559, "learning_rate": 0.0002, "epoch": 3.7895762299074525, "step": 3890}, {"loss": 1.4647, "grad_norm": 0.6706996560096741, "learning_rate": 0.0002, "epoch": 3.799318071115441, "step": 3900}, {"loss": 1.4104, "grad_norm": 0.7026948928833008, "learning_rate": 0.0002, "epoch": 3.809059912323429, "step": 3910}, {"loss": 1.5487, "grad_norm": 0.6437768340110779, "learning_rate": 0.0002, "epoch": 3.8188017535314174, "step": 3920}, {"loss": 1.4678, "grad_norm": 0.7015706300735474, "learning_rate": 0.0002, "epoch": 3.828543594739406, "step": 3930}, {"loss": 1.4891, "grad_norm": 0.7049482464790344, "learning_rate": 0.0002, "epoch": 3.838285435947394, "step": 3940}, {"loss": 1.4208, "grad_norm": 0.6533724665641785, "learning_rate": 0.0002, "epoch": 3.8480272771553823, "step": 3950}, {"loss": 1.4435, "grad_norm": 0.7312499284744263, "learning_rate": 0.0002, "epoch": 3.857769118363371, "step": 3960}, {"loss": 1.3886, "grad_norm": 0.6858801245689392, "learning_rate": 0.0002, "epoch": 3.867510959571359, "step": 3970}, {"loss": 1.4423, "grad_norm": 0.770423173904419, "learning_rate": 0.0002, "epoch": 3.877252800779347, "step": 3980}, {"loss": 1.5029, "grad_norm": 0.6987539529800415, "learning_rate": 0.0002, "epoch": 3.886994641987336, "step": 3990}, {"loss": 1.4791, "grad_norm": 0.7072722315788269, "learning_rate": 0.0002, "epoch": 3.896736483195324, "step": 4000}, {"loss": 1.528, "grad_norm": 0.6492931842803955, "learning_rate": 0.0002, "epoch": 3.906478324403312, "step": 4010}, {"loss": 1.3824, "grad_norm": 0.7716232538223267, "learning_rate": 0.0002, "epoch": 3.9162201656113007, "step": 4020}, {"loss": 1.4758, "grad_norm": 0.722949743270874, "learning_rate": 0.0002, "epoch": 3.925962006819289, "step": 4030}, {"loss": 1.3914, "grad_norm": 0.7434365749359131, "learning_rate": 0.0002, "epoch": 3.935703848027277, "step": 4040}, {"loss": 1.4763, "grad_norm": 0.6691509485244751, "learning_rate": 0.0002, "epoch": 3.9454456892352656, "step": 4050}, {"loss": 1.4555, "grad_norm": 0.6850284337997437, "learning_rate": 0.0002, "epoch": 3.9551875304432538, "step": 4060}, {"loss": 1.5275, "grad_norm": 0.6954452991485596, "learning_rate": 0.0002, "epoch": 3.964929371651242, "step": 4070}, {"loss": 1.417, "grad_norm": 0.9316364526748657, "learning_rate": 0.0002, "epoch": 3.9746712128592305, "step": 4080}, {"loss": 1.4532, "grad_norm": 0.6908289194107056, "learning_rate": 0.0002, "epoch": 3.9844130540672187, "step": 4090}, {"loss": 1.4404, "grad_norm": 0.666782021522522, "learning_rate": 0.0002, "epoch": 3.994154895275207, "step": 4100}, {"eval_loss": 1.9233275651931763, "eval_runtime": 55.9536, "eval_samples_per_second": 9.061, "eval_steps_per_second": 1.144, "epoch": 4.0, "step": 4106}, {"loss": 1.3489, "grad_norm": 0.7726166248321533, "learning_rate": 0.0002, "epoch": 4.003896736483195, "step": 4110}, {"loss": 1.1415, "grad_norm": 1.1338967084884644, "learning_rate": 0.0002, "epoch": 4.013638577691184, "step": 4120}, {"loss": 1.2212, "grad_norm": 0.9530029296875, "learning_rate": 0.0002, "epoch": 4.023380418899172, "step": 4130}, {"loss": 1.2002, "grad_norm": 1.1058554649353027, "learning_rate": 0.0002, "epoch": 4.03312226010716, "step": 4140}, {"loss": 1.2381, "grad_norm": 0.8765049576759338, "learning_rate": 0.0002, "epoch": 4.042864101315149, "step": 4150}, {"loss": 1.2708, "grad_norm": 1.1774667501449585, "learning_rate": 0.0002, "epoch": 4.052605942523137, "step": 4160}, {"loss": 1.2116, "grad_norm": 0.9301433563232422, "learning_rate": 0.0002, "epoch": 4.062347783731125, "step": 4170}, {"loss": 1.1807, "grad_norm": 1.0196778774261475, "learning_rate": 0.0002, "epoch": 4.072089624939114, "step": 4180}, {"loss": 1.2602, "grad_norm": 1.1380577087402344, "learning_rate": 0.0002, "epoch": 4.081831466147102, "step": 4190}, {"loss": 1.2521, "grad_norm": 0.9121319651603699, "learning_rate": 0.0002, "epoch": 4.09157330735509, "step": 4200}, {"loss": 1.1747, "grad_norm": 0.9495378732681274, "learning_rate": 0.0002, "epoch": 4.101315148563079, "step": 4210}, {"loss": 1.1829, "grad_norm": 0.8058680295944214, "learning_rate": 0.0002, "epoch": 4.1110569897710665, "step": 4220}, {"loss": 1.1732, "grad_norm": 1.000887393951416, "learning_rate": 0.0002, "epoch": 4.120798830979055, "step": 4230}, {"loss": 1.1947, "grad_norm": 0.9529102444648743, "learning_rate": 0.0002, "epoch": 4.130540672187044, "step": 4240}, {"loss": 1.2104, "grad_norm": 1.0257115364074707, "learning_rate": 0.0002, "epoch": 4.140282513395031, "step": 4250}, {"loss": 1.2293, "grad_norm": 0.9590303897857666, "learning_rate": 0.0002, "epoch": 4.15002435460302, "step": 4260}, {"loss": 1.1918, "grad_norm": 1.065291166305542, "learning_rate": 0.0002, "epoch": 4.159766195811009, "step": 4270}, {"loss": 1.2323, "grad_norm": 0.8819697499275208, "learning_rate": 0.0002, "epoch": 4.169508037018996, "step": 4280}, {"loss": 1.2167, "grad_norm": 1.0335261821746826, "learning_rate": 0.0002, "epoch": 4.179249878226985, "step": 4290}, {"loss": 1.2131, "grad_norm": 0.8872809410095215, "learning_rate": 0.0002, "epoch": 4.1889917194349735, "step": 4300}, {"loss": 1.2794, "grad_norm": 0.9883159399032593, "learning_rate": 0.0002, "epoch": 4.198733560642961, "step": 4310}, {"loss": 1.2544, "grad_norm": 1.0254192352294922, "learning_rate": 0.0002, "epoch": 4.20847540185095, "step": 4320}, {"loss": 1.2595, "grad_norm": 0.9432600736618042, "learning_rate": 0.0002, "epoch": 4.218217243058938, "step": 4330}, {"loss": 1.2684, "grad_norm": 1.1008676290512085, "learning_rate": 0.0002, "epoch": 4.227959084266926, "step": 4340}, {"loss": 1.2149, "grad_norm": 1.0829699039459229, "learning_rate": 0.0002, "epoch": 4.237700925474915, "step": 4350}, {"loss": 1.2621, "grad_norm": 1.016847848892212, "learning_rate": 0.0002, "epoch": 4.247442766682903, "step": 4360}, {"loss": 1.2375, "grad_norm": 0.8924864530563354, "learning_rate": 0.0002, "epoch": 4.257184607890891, "step": 4370}, {"loss": 1.1987, "grad_norm": 0.9300530552864075, "learning_rate": 0.0002, "epoch": 4.26692644909888, "step": 4380}, {"loss": 1.1696, "grad_norm": 0.9684814810752869, "learning_rate": 0.0002, "epoch": 4.276668290306868, "step": 4390}, {"loss": 1.2006, "grad_norm": 0.9916250705718994, "learning_rate": 0.0002, "epoch": 4.286410131514856, "step": 4400}, {"loss": 1.2402, "grad_norm": 0.903680145740509, "learning_rate": 0.0002, "epoch": 4.2961519727228445, "step": 4410}, {"loss": 1.2022, "grad_norm": 0.8713505268096924, "learning_rate": 0.0002, "epoch": 4.305893813930833, "step": 4420}, {"loss": 1.1957, "grad_norm": 0.9983905553817749, "learning_rate": 0.0002, "epoch": 4.315635655138821, "step": 4430}, {"loss": 1.2676, "grad_norm": 1.1689040660858154, "learning_rate": 0.0002, "epoch": 4.3253774963468095, "step": 4440}, {"loss": 1.2166, "grad_norm": 0.9316853880882263, "learning_rate": 0.0002, "epoch": 4.335119337554798, "step": 4450}, {"loss": 1.222, "grad_norm": 0.9175887107849121, "learning_rate": 0.0002, "epoch": 4.344861178762786, "step": 4460}, {"loss": 1.2571, "grad_norm": 0.9348906874656677, "learning_rate": 0.0002, "epoch": 4.354603019970774, "step": 4470}, {"loss": 1.2764, "grad_norm": 0.9727016687393188, "learning_rate": 0.0002, "epoch": 4.364344861178763, "step": 4480}, {"loss": 1.2616, "grad_norm": 0.9843429923057556, "learning_rate": 0.0002, "epoch": 4.374086702386751, "step": 4490}, {"loss": 1.2488, "grad_norm": 0.9615852236747742, "learning_rate": 0.0002, "epoch": 4.383828543594739, "step": 4500}, {"loss": 1.1718, "grad_norm": 0.9688583612442017, "learning_rate": 0.0002, "epoch": 4.393570384802728, "step": 4510}, {"loss": 1.2546, "grad_norm": 0.9933668375015259, "learning_rate": 0.0002, "epoch": 4.403312226010716, "step": 4520}, {"loss": 1.2355, "grad_norm": 1.0626686811447144, "learning_rate": 0.0002, "epoch": 4.413054067218704, "step": 4530}, {"loss": 1.2425, "grad_norm": 0.9536267518997192, "learning_rate": 0.0002, "epoch": 4.422795908426693, "step": 4540}, {"loss": 1.2562, "grad_norm": 0.9777140021324158, "learning_rate": 0.0002, "epoch": 4.432537749634681, "step": 4550}, {"loss": 1.2878, "grad_norm": 0.980780839920044, "learning_rate": 0.0002, "epoch": 4.442279590842669, "step": 4560}, {"loss": 1.2597, "grad_norm": 1.0147196054458618, "learning_rate": 0.0002, "epoch": 4.452021432050658, "step": 4570}, {"loss": 1.2148, "grad_norm": 0.9763361811637878, "learning_rate": 0.0002, "epoch": 4.461763273258645, "step": 4580}, {"loss": 1.3076, "grad_norm": 1.0300798416137695, "learning_rate": 0.0002, "epoch": 4.471505114466634, "step": 4590}, {"loss": 1.2665, "grad_norm": 0.8833121657371521, "learning_rate": 0.0002, "epoch": 4.481246955674623, "step": 4600}, {"loss": 1.1899, "grad_norm": 1.1214020252227783, "learning_rate": 0.0002, "epoch": 4.490988796882611, "step": 4610}, {"loss": 1.2579, "grad_norm": 0.8843787908554077, "learning_rate": 0.0002, "epoch": 4.500730638090599, "step": 4620}, {"loss": 1.2633, "grad_norm": 0.9942020773887634, "learning_rate": 0.0002, "epoch": 4.5104724792985875, "step": 4630}, {"loss": 1.3172, "grad_norm": 1.0033202171325684, "learning_rate": 0.0002, "epoch": 4.520214320506576, "step": 4640}, {"loss": 1.2024, "grad_norm": 0.8767235279083252, "learning_rate": 0.0002, "epoch": 4.529956161714564, "step": 4650}, {"loss": 1.2714, "grad_norm": 1.0117276906967163, "learning_rate": 0.0002, "epoch": 4.539698002922552, "step": 4660}, {"loss": 1.2911, "grad_norm": 1.2787362337112427, "learning_rate": 0.0002, "epoch": 4.549439844130541, "step": 4670}, {"loss": 1.2603, "grad_norm": 0.8824878931045532, "learning_rate": 0.0002, "epoch": 4.559181685338529, "step": 4680}, {"loss": 1.2905, "grad_norm": 0.9209560751914978, "learning_rate": 0.0002, "epoch": 4.568923526546517, "step": 4690}, {"loss": 1.1916, "grad_norm": 1.1064010858535767, "learning_rate": 0.0002, "epoch": 4.578665367754506, "step": 4700}, {"loss": 1.2217, "grad_norm": 0.8914572596549988, "learning_rate": 0.0002, "epoch": 4.588407208962494, "step": 4710}, {"loss": 1.2861, "grad_norm": 1.0412265062332153, "learning_rate": 0.0002, "epoch": 4.598149050170482, "step": 4720}, {"loss": 1.262, "grad_norm": 1.1950221061706543, "learning_rate": 0.0002, "epoch": 4.607890891378471, "step": 4730}, {"loss": 1.2659, "grad_norm": 0.8938062787055969, "learning_rate": 0.0002, "epoch": 4.617632732586459, "step": 4740}, {"loss": 1.2621, "grad_norm": 0.9849569201469421, "learning_rate": 0.0002, "epoch": 4.627374573794447, "step": 4750}, {"loss": 1.2341, "grad_norm": 1.0081515312194824, "learning_rate": 0.0002, "epoch": 4.637116415002436, "step": 4760}, {"loss": 1.2023, "grad_norm": 0.8566309213638306, "learning_rate": 0.0002, "epoch": 4.6468582562104235, "step": 4770}, {"loss": 1.2723, "grad_norm": 1.1750118732452393, "learning_rate": 0.0002, "epoch": 4.656600097418412, "step": 4780}, {"loss": 1.2537, "grad_norm": 0.925502598285675, "learning_rate": 0.0002, "epoch": 4.666341938626401, "step": 4790}, {"loss": 1.2146, "grad_norm": 1.0402472019195557, "learning_rate": 0.0002, "epoch": 4.676083779834388, "step": 4800}, {"loss": 1.2555, "grad_norm": 0.9772472977638245, "learning_rate": 0.0002, "epoch": 4.685825621042377, "step": 4810}, {"loss": 1.2667, "grad_norm": 0.9082779288291931, "learning_rate": 0.0002, "epoch": 4.695567462250366, "step": 4820}, {"loss": 1.2465, "grad_norm": 0.8026862740516663, "learning_rate": 0.0002, "epoch": 4.705309303458353, "step": 4830}, {"loss": 1.3369, "grad_norm": 1.1631089448928833, "learning_rate": 0.0002, "epoch": 4.715051144666342, "step": 4840}, {"loss": 1.261, "grad_norm": 0.9384787678718567, "learning_rate": 0.0002, "epoch": 4.7247929858743305, "step": 4850}, {"loss": 1.2588, "grad_norm": 1.2151581048965454, "learning_rate": 0.0002, "epoch": 4.734534827082318, "step": 4860}, {"loss": 1.363, "grad_norm": 0.9679436087608337, "learning_rate": 0.0002, "epoch": 4.744276668290307, "step": 4870}, {"loss": 1.3292, "grad_norm": 0.8352158069610596, "learning_rate": 0.0002, "epoch": 4.754018509498295, "step": 4880}, {"loss": 1.3056, "grad_norm": 1.0205804109573364, "learning_rate": 0.0002, "epoch": 4.763760350706283, "step": 4890}, {"loss": 1.223, "grad_norm": 0.9814772605895996, "learning_rate": 0.0002, "epoch": 4.773502191914272, "step": 4900}, {"loss": 1.3114, "grad_norm": 1.002854347229004, "learning_rate": 0.0002, "epoch": 4.78324403312226, "step": 4910}, {"loss": 1.3143, "grad_norm": 1.1609505414962769, "learning_rate": 0.0002, "epoch": 4.792985874330248, "step": 4920}, {"loss": 1.3166, "grad_norm": 0.9354982376098633, "learning_rate": 0.0002, "epoch": 4.802727715538237, "step": 4930}, {"loss": 1.2978, "grad_norm": 0.9761685729026794, "learning_rate": 0.0002, "epoch": 4.812469556746225, "step": 4940}, {"loss": 1.2709, "grad_norm": 1.0604596138000488, "learning_rate": 0.0002, "epoch": 4.822211397954213, "step": 4950}, {"loss": 1.2765, "grad_norm": 1.0902808904647827, "learning_rate": 0.0002, "epoch": 4.8319532391622015, "step": 4960}, {"loss": 1.3073, "grad_norm": 1.0174955129623413, "learning_rate": 0.0002, "epoch": 4.84169508037019, "step": 4970}, {"loss": 1.3141, "grad_norm": 1.0995253324508667, "learning_rate": 0.0002, "epoch": 4.851436921578179, "step": 4980}, {"loss": 1.3006, "grad_norm": 0.880993127822876, "learning_rate": 0.0002, "epoch": 4.8611787627861665, "step": 4990}, {"loss": 1.2547, "grad_norm": 0.9472237825393677, "learning_rate": 0.0002, "epoch": 4.870920603994155, "step": 5000}, {"loss": 1.4078, "grad_norm": 0.9504236578941345, "learning_rate": 0.0002, "epoch": 4.880662445202143, "step": 5010}, {"loss": 1.2791, "grad_norm": 1.1261742115020752, "learning_rate": 0.0002, "epoch": 4.890404286410131, "step": 5020}, {"loss": 1.3707, "grad_norm": 0.904674768447876, "learning_rate": 0.0002, "epoch": 4.90014612761812, "step": 5030}, {"loss": 1.2762, "grad_norm": 0.8828991055488586, "learning_rate": 0.0002, "epoch": 4.909887968826109, "step": 5040}, {"loss": 1.2905, "grad_norm": 1.0156532526016235, "learning_rate": 0.0002, "epoch": 4.919629810034096, "step": 5050}, {"loss": 1.3079, "grad_norm": 0.8975168466567993, "learning_rate": 0.0002, "epoch": 4.929371651242085, "step": 5060}, {"loss": 1.3322, "grad_norm": 0.9787213802337646, "learning_rate": 0.0002, "epoch": 4.939113492450073, "step": 5070}, {"loss": 1.2533, "grad_norm": 1.0801568031311035, "learning_rate": 0.0002, "epoch": 4.948855333658061, "step": 5080}, {"loss": 1.238, "grad_norm": 1.0655089616775513, "learning_rate": 0.0002, "epoch": 4.95859717486605, "step": 5090}, {"loss": 1.2449, "grad_norm": 0.8941320180892944, "learning_rate": 0.0002, "epoch": 4.968339016074038, "step": 5100}, {"loss": 1.2846, "grad_norm": 1.050621747970581, "learning_rate": 0.0002, "epoch": 4.978080857282026, "step": 5110}, {"loss": 1.3791, "grad_norm": 0.9724781513214111, "learning_rate": 0.0002, "epoch": 4.987822698490015, "step": 5120}, {"loss": 1.292, "grad_norm": 0.9850538969039917, "learning_rate": 0.0002, "epoch": 4.997564539698003, "step": 5130}, {"eval_loss": 2.0824170112609863, "eval_runtime": 55.592, "eval_samples_per_second": 9.12, "eval_steps_per_second": 1.151, "epoch": 4.9995129079396, "step": 5132}, {"loss": 1.037, "grad_norm": 1.0096189975738525, "learning_rate": 0.0002, "epoch": 5.007306380905991, "step": 5140}, {"loss": 1.0003, "grad_norm": 1.2403408288955688, "learning_rate": 0.0002, "epoch": 5.01704822211398, "step": 5150}, {"loss": 1.0129, "grad_norm": 1.1243221759796143, "learning_rate": 0.0002, "epoch": 5.026790063321968, "step": 5160}, {"loss": 0.9815, "grad_norm": 1.4745502471923828, "learning_rate": 0.0002, "epoch": 5.036531904529956, "step": 5170}, {"loss": 0.9715, "grad_norm": 1.1913198232650757, "learning_rate": 0.0002, "epoch": 5.0462737457379445, "step": 5180}, {"loss": 0.9282, "grad_norm": 1.2732855081558228, "learning_rate": 0.0002, "epoch": 5.056015586945933, "step": 5190}, {"loss": 0.9857, "grad_norm": 1.1737396717071533, "learning_rate": 0.0002, "epoch": 5.065757428153921, "step": 5200}, {"loss": 0.9754, "grad_norm": 1.4162768125534058, "learning_rate": 0.0002, "epoch": 5.075499269361909, "step": 5210}, {"loss": 1.0333, "grad_norm": 1.528274655342102, "learning_rate": 0.0002, "epoch": 5.085241110569898, "step": 5220}, {"loss": 1.0227, "grad_norm": 1.3966618776321411, "learning_rate": 0.0002, "epoch": 5.094982951777886, "step": 5230}, {"loss": 0.987, "grad_norm": 1.3427953720092773, "learning_rate": 0.0002, "epoch": 5.104724792985874, "step": 5240}, {"loss": 1.0353, "grad_norm": 1.6533905267715454, "learning_rate": 0.0002, "epoch": 5.114466634193863, "step": 5250}, {"loss": 1.0452, "grad_norm": 1.4114865064620972, "learning_rate": 0.0002, "epoch": 5.124208475401851, "step": 5260}, {"loss": 1.067, "grad_norm": 1.5460708141326904, "learning_rate": 0.0002, "epoch": 5.133950316609839, "step": 5270}, {"loss": 1.0667, "grad_norm": 1.3491919040679932, "learning_rate": 0.0002, "epoch": 5.143692157817828, "step": 5280}, {"loss": 0.9957, "grad_norm": 1.2208969593048096, "learning_rate": 0.0002, "epoch": 5.153433999025816, "step": 5290}, {"loss": 1.0362, "grad_norm": 1.1141403913497925, "learning_rate": 0.0002, "epoch": 5.163175840233804, "step": 5300}, {"loss": 0.9744, "grad_norm": 1.2938064336776733, "learning_rate": 0.0002, "epoch": 5.172917681441793, "step": 5310}, {"loss": 1.0438, "grad_norm": 1.2704918384552002, "learning_rate": 0.0002, "epoch": 5.1826595226497805, "step": 5320}, {"loss": 1.0015, "grad_norm": 1.3928544521331787, "learning_rate": 0.0002, "epoch": 5.192401363857769, "step": 5330}, {"loss": 1.025, "grad_norm": 1.1993824243545532, "learning_rate": 0.0002, "epoch": 5.202143205065758, "step": 5340}, {"loss": 1.0195, "grad_norm": 1.5913670063018799, "learning_rate": 0.0002, "epoch": 5.211885046273745, "step": 5350}, {"loss": 1.0113, "grad_norm": 1.1577855348587036, "learning_rate": 0.0002, "epoch": 5.221626887481734, "step": 5360}, {"loss": 1.0684, "grad_norm": 1.4535993337631226, "learning_rate": 0.0002, "epoch": 5.231368728689723, "step": 5370}, {"loss": 1.0255, "grad_norm": 1.5068976879119873, "learning_rate": 0.0002, "epoch": 5.24111056989771, "step": 5380}, {"loss": 1.0068, "grad_norm": 1.2365459203720093, "learning_rate": 0.0002, "epoch": 5.250852411105699, "step": 5390}, {"loss": 1.0145, "grad_norm": 1.3197922706604004, "learning_rate": 0.0002, "epoch": 5.2605942523136875, "step": 5400}, {"loss": 1.0767, "grad_norm": 1.2395117282867432, "learning_rate": 0.0002, "epoch": 5.270336093521675, "step": 5410}, {"loss": 1.0292, "grad_norm": 1.1841236352920532, "learning_rate": 0.0002, "epoch": 5.280077934729664, "step": 5420}, {"loss": 1.0233, "grad_norm": 1.218003749847412, "learning_rate": 0.0002, "epoch": 5.289819775937652, "step": 5430}, {"loss": 1.0093, "grad_norm": 1.2210947275161743, "learning_rate": 0.0002, "epoch": 5.29956161714564, "step": 5440}, {"loss": 0.9619, "grad_norm": 1.266006588935852, "learning_rate": 0.0002, "epoch": 5.309303458353629, "step": 5450}, {"loss": 1.0352, "grad_norm": 1.2598075866699219, "learning_rate": 0.0002, "epoch": 5.319045299561617, "step": 5460}, {"loss": 1.0929, "grad_norm": 1.2410019636154175, "learning_rate": 0.0002, "epoch": 5.328787140769606, "step": 5470}, {"loss": 1.058, "grad_norm": 1.249698519706726, "learning_rate": 0.0002, "epoch": 5.338528981977594, "step": 5480}, {"loss": 1.0457, "grad_norm": 1.2398173809051514, "learning_rate": 0.0002, "epoch": 5.348270823185582, "step": 5490}, {"loss": 1.0139, "grad_norm": 1.2416654825210571, "learning_rate": 0.0002, "epoch": 5.35801266439357, "step": 5500}, {"loss": 1.0609, "grad_norm": 1.398706316947937, "learning_rate": 0.0002, "epoch": 5.3677545056015585, "step": 5510}, {"loss": 1.0512, "grad_norm": 1.3049418926239014, "learning_rate": 0.0002, "epoch": 5.377496346809547, "step": 5520}, {"loss": 1.0912, "grad_norm": 1.2528893947601318, "learning_rate": 0.0002, "epoch": 5.387238188017536, "step": 5530}, {"loss": 1.0619, "grad_norm": 1.2963255643844604, "learning_rate": 0.0002, "epoch": 5.3969800292255234, "step": 5540}, {"loss": 1.0194, "grad_norm": 1.494231104850769, "learning_rate": 0.0002, "epoch": 5.406721870433512, "step": 5550}, {"loss": 1.0179, "grad_norm": 1.2760992050170898, "learning_rate": 0.0002, "epoch": 5.416463711641501, "step": 5560}, {"loss": 1.1088, "grad_norm": 1.195292592048645, "learning_rate": 0.0002, "epoch": 5.426205552849488, "step": 5570}, {"loss": 1.0859, "grad_norm": 1.6408965587615967, "learning_rate": 0.0002, "epoch": 5.435947394057477, "step": 5580}, {"loss": 1.0868, "grad_norm": 1.3092058897018433, "learning_rate": 0.0002, "epoch": 5.4456892352654656, "step": 5590}, {"loss": 1.006, "grad_norm": 1.2960586547851562, "learning_rate": 0.0002, "epoch": 5.455431076473453, "step": 5600}, {"loss": 1.0257, "grad_norm": 1.3560487031936646, "learning_rate": 0.0002, "epoch": 5.465172917681442, "step": 5610}, {"loss": 1.0314, "grad_norm": 1.1896311044692993, "learning_rate": 0.0002, "epoch": 5.4749147588894305, "step": 5620}, {"loss": 1.0435, "grad_norm": 1.3145595788955688, "learning_rate": 0.0002, "epoch": 5.484656600097418, "step": 5630}, {"loss": 1.0456, "grad_norm": 1.2207404375076294, "learning_rate": 0.0002, "epoch": 5.494398441305407, "step": 5640}, {"loss": 1.0823, "grad_norm": 1.266015887260437, "learning_rate": 0.0002, "epoch": 5.504140282513395, "step": 5650}, {"loss": 1.0696, "grad_norm": 1.2478289604187012, "learning_rate": 0.0002, "epoch": 5.513882123721383, "step": 5660}, {"loss": 1.0695, "grad_norm": 1.4851372241973877, "learning_rate": 0.0002, "epoch": 5.523623964929372, "step": 5670}, {"loss": 1.0736, "grad_norm": 1.4478679895401, "learning_rate": 0.0002, "epoch": 5.53336580613736, "step": 5680}, {"loss": 1.043, "grad_norm": 1.1079537868499756, "learning_rate": 0.0002, "epoch": 5.543107647345348, "step": 5690}, {"loss": 1.1107, "grad_norm": 1.4201879501342773, "learning_rate": 0.0002, "epoch": 5.552849488553337, "step": 5700}, {"loss": 1.0697, "grad_norm": 1.2092000246047974, "learning_rate": 0.0002, "epoch": 5.562591329761325, "step": 5710}, {"loss": 0.9868, "grad_norm": 1.4515851736068726, "learning_rate": 0.0002, "epoch": 5.572333170969313, "step": 5720}, {"loss": 1.1547, "grad_norm": 1.3260412216186523, "learning_rate": 0.0002, "epoch": 5.5820750121773015, "step": 5730}, {"loss": 1.1388, "grad_norm": 1.248191475868225, "learning_rate": 0.0002, "epoch": 5.59181685338529, "step": 5740}, {"loss": 1.0597, "grad_norm": 1.2037307024002075, "learning_rate": 0.0002, "epoch": 5.601558694593278, "step": 5750}, {"loss": 1.1425, "grad_norm": 1.341237187385559, "learning_rate": 0.0002, "epoch": 5.611300535801266, "step": 5760}, {"loss": 1.0942, "grad_norm": 1.130115270614624, "learning_rate": 0.0002, "epoch": 5.621042377009255, "step": 5770}, {"loss": 1.1029, "grad_norm": 1.3834772109985352, "learning_rate": 0.0002, "epoch": 5.630784218217243, "step": 5780}, {"loss": 1.0825, "grad_norm": 1.2586270570755005, "learning_rate": 0.0002, "epoch": 5.640526059425231, "step": 5790}, {"loss": 1.0186, "grad_norm": 1.3233023881912231, "learning_rate": 0.0002, "epoch": 5.65026790063322, "step": 5800}, {"loss": 1.0557, "grad_norm": 1.2711341381072998, "learning_rate": 0.0002, "epoch": 5.660009741841208, "step": 5810}, {"loss": 1.0897, "grad_norm": 1.3867720365524292, "learning_rate": 0.0002, "epoch": 5.669751583049196, "step": 5820}, {"loss": 1.0776, "grad_norm": 1.4783269166946411, "learning_rate": 0.0002, "epoch": 5.679493424257185, "step": 5830}, {"loss": 1.0632, "grad_norm": 1.2744768857955933, "learning_rate": 0.0002, "epoch": 5.6892352654651726, "step": 5840}, {"loss": 1.1484, "grad_norm": 1.3405882120132446, "learning_rate": 0.0002, "epoch": 5.698977106673161, "step": 5850}, {"loss": 1.0975, "grad_norm": 1.204300880432129, "learning_rate": 0.0002, "epoch": 5.70871894788115, "step": 5860}, {"loss": 1.0494, "grad_norm": 1.2954572439193726, "learning_rate": 0.0002, "epoch": 5.7184607890891375, "step": 5870}, {"loss": 1.0643, "grad_norm": 1.5478382110595703, "learning_rate": 0.0002, "epoch": 5.728202630297126, "step": 5880}, {"loss": 1.0582, "grad_norm": 1.2095842361450195, "learning_rate": 0.0002, "epoch": 5.737944471505115, "step": 5890}, {"loss": 1.1, "grad_norm": 1.0691519975662231, "learning_rate": 0.0002, "epoch": 5.747686312713103, "step": 5900}, {"loss": 1.0906, "grad_norm": 1.1920677423477173, "learning_rate": 0.0002, "epoch": 5.757428153921091, "step": 5910}, {"loss": 1.1746, "grad_norm": 1.2051277160644531, "learning_rate": 0.0002, "epoch": 5.76716999512908, "step": 5920}, {"loss": 1.1221, "grad_norm": 1.197490930557251, "learning_rate": 0.0002, "epoch": 5.776911836337067, "step": 5930}, {"loss": 1.07, "grad_norm": 1.2003998756408691, "learning_rate": 0.0002, "epoch": 5.786653677545056, "step": 5940}, {"loss": 1.0938, "grad_norm": 1.2323646545410156, "learning_rate": 0.0002, "epoch": 5.7963955187530445, "step": 5950}, {"loss": 1.1443, "grad_norm": 1.2593932151794434, "learning_rate": 0.0002, "epoch": 5.806137359961033, "step": 5960}, {"loss": 1.0829, "grad_norm": 1.1835976839065552, "learning_rate": 0.0002, "epoch": 5.815879201169021, "step": 5970}, {"loss": 1.1056, "grad_norm": 1.4770104885101318, "learning_rate": 0.0002, "epoch": 5.825621042377009, "step": 5980}, {"loss": 1.1934, "grad_norm": 1.1025809049606323, "learning_rate": 0.0002, "epoch": 5.835362883584997, "step": 5990}, {"loss": 1.1323, "grad_norm": 1.364588975906372, "learning_rate": 0.0002, "epoch": 5.845104724792986, "step": 6000}, {"loss": 1.1234, "grad_norm": 1.2340112924575806, "learning_rate": 0.0002, "epoch": 5.854846566000974, "step": 6010}, {"loss": 1.1123, "grad_norm": 1.4925711154937744, "learning_rate": 0.0002, "epoch": 5.864588407208963, "step": 6020}, {"loss": 1.12, "grad_norm": 1.3516744375228882, "learning_rate": 0.0002, "epoch": 5.874330248416951, "step": 6030}, {"loss": 1.1399, "grad_norm": 1.2058138847351074, "learning_rate": 0.0002, "epoch": 5.884072089624939, "step": 6040}, {"loss": 1.1074, "grad_norm": 1.13870108127594, "learning_rate": 0.0002, "epoch": 5.893813930832927, "step": 6050}, {"loss": 1.088, "grad_norm": 1.1587319374084473, "learning_rate": 0.0002, "epoch": 5.9035557720409155, "step": 6060}, {"loss": 1.1376, "grad_norm": 1.164481520652771, "learning_rate": 0.0002, "epoch": 5.913297613248904, "step": 6070}, {"loss": 1.1262, "grad_norm": 1.2115206718444824, "learning_rate": 0.0002, "epoch": 5.923039454456893, "step": 6080}, {"loss": 1.1345, "grad_norm": 1.3201590776443481, "learning_rate": 0.0002, "epoch": 5.93278129566488, "step": 6090}, {"loss": 1.1288, "grad_norm": 1.287380576133728, "learning_rate": 0.0002, "epoch": 5.942523136872869, "step": 6100}, {"loss": 1.1475, "grad_norm": 1.1820166110992432, "learning_rate": 0.0002, "epoch": 5.952264978080858, "step": 6110}, {"loss": 1.1112, "grad_norm": 1.2550667524337769, "learning_rate": 0.0002, "epoch": 5.962006819288845, "step": 6120}, {"loss": 1.1528, "grad_norm": 1.3547813892364502, "learning_rate": 0.0002, "epoch": 5.971748660496834, "step": 6130}, {"loss": 1.0557, "grad_norm": 1.260842204093933, "learning_rate": 0.0002, "epoch": 5.9814905017048225, "step": 6140}, {"loss": 1.1119, "grad_norm": 1.1643036603927612, "learning_rate": 0.0002, "epoch": 5.99123234291281, "step": 6150}]}
+{"epoch": 6.9995129079396, "step": 7185, "epoch_duration": 1628.9270315170288, "total_accumulated_duration": 12510.417803525925, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}, {"eval_loss": 1.8463934659957886, "eval_runtime": 56.2401, "eval_samples_per_second": 9.015, "eval_steps_per_second": 1.138, "epoch": 2.9995129079396006, "step": 3079}, {"loss": 1.5744, "grad_norm": 0.49992576241493225, "learning_rate": 0.0002, "epoch": 3.0004870920603994, "step": 3080}, {"loss": 1.4125, "grad_norm": 0.8242783546447754, "learning_rate": 0.0002, "epoch": 3.0102289332683876, "step": 3090}, {"loss": 1.394, "grad_norm": 0.6330569386482239, "learning_rate": 0.0002, "epoch": 3.019970774476376, "step": 3100}, {"loss": 1.4942, "grad_norm": 0.566097617149353, "learning_rate": 0.0002, "epoch": 3.0297126156843643, "step": 3110}, {"loss": 1.4365, "grad_norm": 0.6337586045265198, "learning_rate": 0.0002, "epoch": 3.0394544568923525, "step": 3120}, {"loss": 1.3916, "grad_norm": 0.7339403033256531, "learning_rate": 0.0002, "epoch": 3.049196298100341, "step": 3130}, {"loss": 1.4617, "grad_norm": 0.7187346816062927, "learning_rate": 0.0002, "epoch": 3.0589381393083293, "step": 3140}, {"loss": 1.3453, "grad_norm": 0.7116255760192871, "learning_rate": 0.0002, "epoch": 3.0686799805163174, "step": 3150}, {"loss": 1.4452, "grad_norm": 0.6493807435035706, "learning_rate": 0.0002, "epoch": 3.078421821724306, "step": 3160}, {"loss": 1.351, "grad_norm": 0.6777266263961792, "learning_rate": 0.0002, "epoch": 3.088163662932294, "step": 3170}, {"loss": 1.4362, "grad_norm": 0.6342006325721741, "learning_rate": 0.0002, "epoch": 3.0979055041402823, "step": 3180}, {"loss": 1.4748, "grad_norm": 0.6608964204788208, "learning_rate": 0.0002, "epoch": 3.107647345348271, "step": 3190}, {"loss": 1.375, "grad_norm": 0.7230247259140015, "learning_rate": 0.0002, "epoch": 3.117389186556259, "step": 3200}, {"loss": 1.4049, "grad_norm": 0.650368332862854, "learning_rate": 0.0002, "epoch": 3.1271310277642472, "step": 3210}, {"loss": 1.409, "grad_norm": 0.7319342494010925, "learning_rate": 0.0002, "epoch": 3.136872868972236, "step": 3220}, {"loss": 1.3872, "grad_norm": 0.7159963846206665, "learning_rate": 0.0002, "epoch": 3.146614710180224, "step": 3230}, {"loss": 1.5076, "grad_norm": 0.8905230164527893, "learning_rate": 0.0002, "epoch": 3.156356551388212, "step": 3240}, {"loss": 1.3161, "grad_norm": 0.6920804381370544, "learning_rate": 0.0002, "epoch": 3.1660983925962007, "step": 3250}, {"loss": 1.3786, "grad_norm": 0.6782063841819763, "learning_rate": 0.0002, "epoch": 3.175840233804189, "step": 3260}, {"loss": 1.5153, "grad_norm": 0.735325276851654, "learning_rate": 0.0002, "epoch": 3.1855820750121775, "step": 3270}, {"loss": 1.4027, "grad_norm": 0.6657978296279907, "learning_rate": 0.0002, "epoch": 3.1953239162201656, "step": 3280}, {"loss": 1.3456, "grad_norm": 0.771315336227417, "learning_rate": 0.0002, "epoch": 3.205065757428154, "step": 3290}, {"loss": 1.3236, "grad_norm": 0.6492983102798462, "learning_rate": 0.0002, "epoch": 3.2148075986361424, "step": 3300}, {"loss": 1.4125, "grad_norm": 0.7513770461082458, "learning_rate": 0.0002, "epoch": 3.2245494398441306, "step": 3310}, {"loss": 1.4032, "grad_norm": 0.7091423869132996, "learning_rate": 0.0002, "epoch": 3.2342912810521187, "step": 3320}, {"loss": 1.4585, "grad_norm": 0.6663975119590759, "learning_rate": 0.0002, "epoch": 3.2440331222601073, "step": 3330}, {"loss": 1.3968, "grad_norm": 0.6813122034072876, "learning_rate": 0.0002, "epoch": 3.2537749634680955, "step": 3340}, {"loss": 1.3681, "grad_norm": 0.6602569818496704, "learning_rate": 0.0002, "epoch": 3.2635168046760836, "step": 3350}, {"loss": 1.4533, "grad_norm": 0.718270480632782, "learning_rate": 0.0002, "epoch": 3.2732586458840722, "step": 3360}, {"loss": 1.4076, "grad_norm": 0.6884173154830933, "learning_rate": 0.0002, "epoch": 3.2830004870920604, "step": 3370}, {"loss": 1.4144, "grad_norm": 0.7039775848388672, "learning_rate": 0.0002, "epoch": 3.2927423283000485, "step": 3380}, {"loss": 1.5077, "grad_norm": 0.7444299459457397, "learning_rate": 0.0002, "epoch": 3.302484169508037, "step": 3390}, {"loss": 1.4255, "grad_norm": 0.7187064290046692, "learning_rate": 0.0002, "epoch": 3.3122260107160253, "step": 3400}, {"loss": 1.3684, "grad_norm": 0.599396288394928, "learning_rate": 0.0002, "epoch": 3.3219678519240134, "step": 3410}, {"loss": 1.4819, "grad_norm": 0.7670390009880066, "learning_rate": 0.0002, "epoch": 3.331709693132002, "step": 3420}, {"loss": 1.4411, "grad_norm": 0.6654478311538696, "learning_rate": 0.0002, "epoch": 3.34145153433999, "step": 3430}, {"loss": 1.4257, "grad_norm": 0.6644385457038879, "learning_rate": 0.0002, "epoch": 3.351193375547979, "step": 3440}, {"loss": 1.4508, "grad_norm": 0.6974098086357117, "learning_rate": 0.0002, "epoch": 3.360935216755967, "step": 3450}, {"loss": 1.3807, "grad_norm": 0.7350399494171143, "learning_rate": 0.0002, "epoch": 3.370677057963955, "step": 3460}, {"loss": 1.4176, "grad_norm": 0.714721143245697, "learning_rate": 0.0002, "epoch": 3.3804188991719437, "step": 3470}, {"loss": 1.4325, "grad_norm": 0.7006027698516846, "learning_rate": 0.0002, "epoch": 3.390160740379932, "step": 3480}, {"loss": 1.4888, "grad_norm": 0.6767925024032593, "learning_rate": 0.0002, "epoch": 3.39990258158792, "step": 3490}, {"loss": 1.4116, "grad_norm": 0.6721355319023132, "learning_rate": 0.0002, "epoch": 3.4096444227959086, "step": 3500}, {"loss": 1.443, "grad_norm": 0.6845725178718567, "learning_rate": 0.0002, "epoch": 3.419386264003897, "step": 3510}, {"loss": 1.4832, "grad_norm": 0.6882196664810181, "learning_rate": 0.0002, "epoch": 3.429128105211885, "step": 3520}, {"loss": 1.4962, "grad_norm": 0.7663240432739258, "learning_rate": 0.0002, "epoch": 3.4388699464198735, "step": 3530}, {"loss": 1.4644, "grad_norm": 0.6304219365119934, "learning_rate": 0.0002, "epoch": 3.4486117876278617, "step": 3540}, {"loss": 1.4918, "grad_norm": 0.668678879737854, "learning_rate": 0.0002, "epoch": 3.45835362883585, "step": 3550}, {"loss": 1.4874, "grad_norm": 0.7526912093162537, "learning_rate": 0.0002, "epoch": 3.4680954700438384, "step": 3560}, {"loss": 1.4249, "grad_norm": 1.089495301246643, "learning_rate": 0.0002, "epoch": 3.4778373112518266, "step": 3570}, {"loss": 1.3871, "grad_norm": 0.7282902002334595, "learning_rate": 0.0002, "epoch": 3.4875791524598148, "step": 3580}, {"loss": 1.5077, "grad_norm": 0.6540156602859497, "learning_rate": 0.0002, "epoch": 3.4973209936678034, "step": 3590}, {"loss": 1.4367, "grad_norm": 0.6449568867683411, "learning_rate": 0.0002, "epoch": 3.5070628348757915, "step": 3600}, {"loss": 1.4532, "grad_norm": 0.7262216210365295, "learning_rate": 0.0002, "epoch": 3.5168046760837797, "step": 3610}, {"loss": 1.4374, "grad_norm": 0.6048615574836731, "learning_rate": 0.0002, "epoch": 3.5265465172917683, "step": 3620}, {"loss": 1.3877, "grad_norm": 0.6780537366867065, "learning_rate": 0.0002, "epoch": 3.5362883584997564, "step": 3630}, {"loss": 1.422, "grad_norm": 0.6851925253868103, "learning_rate": 0.0002, "epoch": 3.5460301997077446, "step": 3640}, {"loss": 1.3425, "grad_norm": 0.6530634164810181, "learning_rate": 0.0002, "epoch": 3.555772040915733, "step": 3650}, {"loss": 1.4879, "grad_norm": 0.7193992733955383, "learning_rate": 0.0002, "epoch": 3.5655138821237213, "step": 3660}, {"loss": 1.4555, "grad_norm": 0.767496645450592, "learning_rate": 0.0002, "epoch": 3.5752557233317095, "step": 3670}, {"loss": 1.4824, "grad_norm": 0.6912919282913208, "learning_rate": 0.0002, "epoch": 3.584997564539698, "step": 3680}, {"loss": 1.4497, "grad_norm": 0.7383436560630798, "learning_rate": 0.0002, "epoch": 3.5947394057476862, "step": 3690}, {"loss": 1.4822, "grad_norm": 0.6746662855148315, "learning_rate": 0.0002, "epoch": 3.6044812469556744, "step": 3700}, {"loss": 1.4904, "grad_norm": 0.6885138750076294, "learning_rate": 0.0002, "epoch": 3.614223088163663, "step": 3710}, {"loss": 1.4044, "grad_norm": 0.6694392561912537, "learning_rate": 0.0002, "epoch": 3.623964929371651, "step": 3720}, {"loss": 1.3719, "grad_norm": 0.812358021736145, "learning_rate": 0.0002, "epoch": 3.6337067705796393, "step": 3730}, {"loss": 1.4603, "grad_norm": 0.7267130017280579, "learning_rate": 0.0002, "epoch": 3.643448611787628, "step": 3740}, {"loss": 1.4574, "grad_norm": 0.6958749294281006, "learning_rate": 0.0002, "epoch": 3.653190452995616, "step": 3750}, {"loss": 1.4346, "grad_norm": 0.6805673241615295, "learning_rate": 0.0002, "epoch": 3.6629322942036042, "step": 3760}, {"loss": 1.4338, "grad_norm": 0.7184410095214844, "learning_rate": 0.0002, "epoch": 3.672674135411593, "step": 3770}, {"loss": 1.3935, "grad_norm": 0.7716330289840698, "learning_rate": 0.0002, "epoch": 3.682415976619581, "step": 3780}, {"loss": 1.384, "grad_norm": 0.6675831079483032, "learning_rate": 0.0002, "epoch": 3.6921578178275696, "step": 3790}, {"loss": 1.401, "grad_norm": 0.6480095386505127, "learning_rate": 0.0002, "epoch": 3.7018996590355577, "step": 3800}, {"loss": 1.5303, "grad_norm": 0.6559418439865112, "learning_rate": 0.0002, "epoch": 3.711641500243546, "step": 3810}, {"loss": 1.4341, "grad_norm": 0.6596545577049255, "learning_rate": 0.0002, "epoch": 3.7213833414515345, "step": 3820}, {"loss": 1.4508, "grad_norm": 0.7172950506210327, "learning_rate": 0.0002, "epoch": 3.7311251826595226, "step": 3830}, {"loss": 1.446, "grad_norm": 0.796148419380188, "learning_rate": 0.0002, "epoch": 3.740867023867511, "step": 3840}, {"loss": 1.4992, "grad_norm": 0.6600322723388672, "learning_rate": 0.0002, "epoch": 3.7506088650754994, "step": 3850}, {"loss": 1.4201, "grad_norm": 0.6776387691497803, "learning_rate": 0.0002, "epoch": 3.7603507062834876, "step": 3860}, {"loss": 1.3893, "grad_norm": 0.7768304347991943, "learning_rate": 0.0002, "epoch": 3.770092547491476, "step": 3870}, {"loss": 1.4886, "grad_norm": 1.0579794645309448, "learning_rate": 0.0002, "epoch": 3.7798343886994643, "step": 3880}, {"loss": 1.4556, "grad_norm": 0.6757252812385559, "learning_rate": 0.0002, "epoch": 3.7895762299074525, "step": 3890}, {"loss": 1.4647, "grad_norm": 0.6706996560096741, "learning_rate": 0.0002, "epoch": 3.799318071115441, "step": 3900}, {"loss": 1.4104, "grad_norm": 0.7026948928833008, "learning_rate": 0.0002, "epoch": 3.809059912323429, "step": 3910}, {"loss": 1.5487, "grad_norm": 0.6437768340110779, "learning_rate": 0.0002, "epoch": 3.8188017535314174, "step": 3920}, {"loss": 1.4678, "grad_norm": 0.7015706300735474, "learning_rate": 0.0002, "epoch": 3.828543594739406, "step": 3930}, {"loss": 1.4891, "grad_norm": 0.7049482464790344, "learning_rate": 0.0002, "epoch": 3.838285435947394, "step": 3940}, {"loss": 1.4208, "grad_norm": 0.6533724665641785, "learning_rate": 0.0002, "epoch": 3.8480272771553823, "step": 3950}, {"loss": 1.4435, "grad_norm": 0.7312499284744263, "learning_rate": 0.0002, "epoch": 3.857769118363371, "step": 3960}, {"loss": 1.3886, "grad_norm": 0.6858801245689392, "learning_rate": 0.0002, "epoch": 3.867510959571359, "step": 3970}, {"loss": 1.4423, "grad_norm": 0.770423173904419, "learning_rate": 0.0002, "epoch": 3.877252800779347, "step": 3980}, {"loss": 1.5029, "grad_norm": 0.6987539529800415, "learning_rate": 0.0002, "epoch": 3.886994641987336, "step": 3990}, {"loss": 1.4791, "grad_norm": 0.7072722315788269, "learning_rate": 0.0002, "epoch": 3.896736483195324, "step": 4000}, {"loss": 1.528, "grad_norm": 0.6492931842803955, "learning_rate": 0.0002, "epoch": 3.906478324403312, "step": 4010}, {"loss": 1.3824, "grad_norm": 0.7716232538223267, "learning_rate": 0.0002, "epoch": 3.9162201656113007, "step": 4020}, {"loss": 1.4758, "grad_norm": 0.722949743270874, "learning_rate": 0.0002, "epoch": 3.925962006819289, "step": 4030}, {"loss": 1.3914, "grad_norm": 0.7434365749359131, "learning_rate": 0.0002, "epoch": 3.935703848027277, "step": 4040}, {"loss": 1.4763, "grad_norm": 0.6691509485244751, "learning_rate": 0.0002, "epoch": 3.9454456892352656, "step": 4050}, {"loss": 1.4555, "grad_norm": 0.6850284337997437, "learning_rate": 0.0002, "epoch": 3.9551875304432538, "step": 4060}, {"loss": 1.5275, "grad_norm": 0.6954452991485596, "learning_rate": 0.0002, "epoch": 3.964929371651242, "step": 4070}, {"loss": 1.417, "grad_norm": 0.9316364526748657, "learning_rate": 0.0002, "epoch": 3.9746712128592305, "step": 4080}, {"loss": 1.4532, "grad_norm": 0.6908289194107056, "learning_rate": 0.0002, "epoch": 3.9844130540672187, "step": 4090}, {"loss": 1.4404, "grad_norm": 0.666782021522522, "learning_rate": 0.0002, "epoch": 3.994154895275207, "step": 4100}, {"eval_loss": 1.9233275651931763, "eval_runtime": 55.9536, "eval_samples_per_second": 9.061, "eval_steps_per_second": 1.144, "epoch": 4.0, "step": 4106}, {"loss": 1.3489, "grad_norm": 0.7726166248321533, "learning_rate": 0.0002, "epoch": 4.003896736483195, "step": 4110}, {"loss": 1.1415, "grad_norm": 1.1338967084884644, "learning_rate": 0.0002, "epoch": 4.013638577691184, "step": 4120}, {"loss": 1.2212, "grad_norm": 0.9530029296875, "learning_rate": 0.0002, "epoch": 4.023380418899172, "step": 4130}, {"loss": 1.2002, "grad_norm": 1.1058554649353027, "learning_rate": 0.0002, "epoch": 4.03312226010716, "step": 4140}, {"loss": 1.2381, "grad_norm": 0.8765049576759338, "learning_rate": 0.0002, "epoch": 4.042864101315149, "step": 4150}, {"loss": 1.2708, "grad_norm": 1.1774667501449585, "learning_rate": 0.0002, "epoch": 4.052605942523137, "step": 4160}, {"loss": 1.2116, "grad_norm": 0.9301433563232422, "learning_rate": 0.0002, "epoch": 4.062347783731125, "step": 4170}, {"loss": 1.1807, "grad_norm": 1.0196778774261475, "learning_rate": 0.0002, "epoch": 4.072089624939114, "step": 4180}, {"loss": 1.2602, "grad_norm": 1.1380577087402344, "learning_rate": 0.0002, "epoch": 4.081831466147102, "step": 4190}, {"loss": 1.2521, "grad_norm": 0.9121319651603699, "learning_rate": 0.0002, "epoch": 4.09157330735509, "step": 4200}, {"loss": 1.1747, "grad_norm": 0.9495378732681274, "learning_rate": 0.0002, "epoch": 4.101315148563079, "step": 4210}, {"loss": 1.1829, "grad_norm": 0.8058680295944214, "learning_rate": 0.0002, "epoch": 4.1110569897710665, "step": 4220}, {"loss": 1.1732, "grad_norm": 1.000887393951416, "learning_rate": 0.0002, "epoch": 4.120798830979055, "step": 4230}, {"loss": 1.1947, "grad_norm": 0.9529102444648743, "learning_rate": 0.0002, "epoch": 4.130540672187044, "step": 4240}, {"loss": 1.2104, "grad_norm": 1.0257115364074707, "learning_rate": 0.0002, "epoch": 4.140282513395031, "step": 4250}, {"loss": 1.2293, "grad_norm": 0.9590303897857666, "learning_rate": 0.0002, "epoch": 4.15002435460302, "step": 4260}, {"loss": 1.1918, "grad_norm": 1.065291166305542, "learning_rate": 0.0002, "epoch": 4.159766195811009, "step": 4270}, {"loss": 1.2323, "grad_norm": 0.8819697499275208, "learning_rate": 0.0002, "epoch": 4.169508037018996, "step": 4280}, {"loss": 1.2167, "grad_norm": 1.0335261821746826, "learning_rate": 0.0002, "epoch": 4.179249878226985, "step": 4290}, {"loss": 1.2131, "grad_norm": 0.8872809410095215, "learning_rate": 0.0002, "epoch": 4.1889917194349735, "step": 4300}, {"loss": 1.2794, "grad_norm": 0.9883159399032593, "learning_rate": 0.0002, "epoch": 4.198733560642961, "step": 4310}, {"loss": 1.2544, "grad_norm": 1.0254192352294922, "learning_rate": 0.0002, "epoch": 4.20847540185095, "step": 4320}, {"loss": 1.2595, "grad_norm": 0.9432600736618042, "learning_rate": 0.0002, "epoch": 4.218217243058938, "step": 4330}, {"loss": 1.2684, "grad_norm": 1.1008676290512085, "learning_rate": 0.0002, "epoch": 4.227959084266926, "step": 4340}, {"loss": 1.2149, "grad_norm": 1.0829699039459229, "learning_rate": 0.0002, "epoch": 4.237700925474915, "step": 4350}, {"loss": 1.2621, "grad_norm": 1.016847848892212, "learning_rate": 0.0002, "epoch": 4.247442766682903, "step": 4360}, {"loss": 1.2375, "grad_norm": 0.8924864530563354, "learning_rate": 0.0002, "epoch": 4.257184607890891, "step": 4370}, {"loss": 1.1987, "grad_norm": 0.9300530552864075, "learning_rate": 0.0002, "epoch": 4.26692644909888, "step": 4380}, {"loss": 1.1696, "grad_norm": 0.9684814810752869, "learning_rate": 0.0002, "epoch": 4.276668290306868, "step": 4390}, {"loss": 1.2006, "grad_norm": 0.9916250705718994, "learning_rate": 0.0002, "epoch": 4.286410131514856, "step": 4400}, {"loss": 1.2402, "grad_norm": 0.903680145740509, "learning_rate": 0.0002, "epoch": 4.2961519727228445, "step": 4410}, {"loss": 1.2022, "grad_norm": 0.8713505268096924, "learning_rate": 0.0002, "epoch": 4.305893813930833, "step": 4420}, {"loss": 1.1957, "grad_norm": 0.9983905553817749, "learning_rate": 0.0002, "epoch": 4.315635655138821, "step": 4430}, {"loss": 1.2676, "grad_norm": 1.1689040660858154, "learning_rate": 0.0002, "epoch": 4.3253774963468095, "step": 4440}, {"loss": 1.2166, "grad_norm": 0.9316853880882263, "learning_rate": 0.0002, "epoch": 4.335119337554798, "step": 4450}, {"loss": 1.222, "grad_norm": 0.9175887107849121, "learning_rate": 0.0002, "epoch": 4.344861178762786, "step": 4460}, {"loss": 1.2571, "grad_norm": 0.9348906874656677, "learning_rate": 0.0002, "epoch": 4.354603019970774, "step": 4470}, {"loss": 1.2764, "grad_norm": 0.9727016687393188, "learning_rate": 0.0002, "epoch": 4.364344861178763, "step": 4480}, {"loss": 1.2616, "grad_norm": 0.9843429923057556, "learning_rate": 0.0002, "epoch": 4.374086702386751, "step": 4490}, {"loss": 1.2488, "grad_norm": 0.9615852236747742, "learning_rate": 0.0002, "epoch": 4.383828543594739, "step": 4500}, {"loss": 1.1718, "grad_norm": 0.9688583612442017, "learning_rate": 0.0002, "epoch": 4.393570384802728, "step": 4510}, {"loss": 1.2546, "grad_norm": 0.9933668375015259, "learning_rate": 0.0002, "epoch": 4.403312226010716, "step": 4520}, {"loss": 1.2355, "grad_norm": 1.0626686811447144, "learning_rate": 0.0002, "epoch": 4.413054067218704, "step": 4530}, {"loss": 1.2425, "grad_norm": 0.9536267518997192, "learning_rate": 0.0002, "epoch": 4.422795908426693, "step": 4540}, {"loss": 1.2562, "grad_norm": 0.9777140021324158, "learning_rate": 0.0002, "epoch": 4.432537749634681, "step": 4550}, {"loss": 1.2878, "grad_norm": 0.980780839920044, "learning_rate": 0.0002, "epoch": 4.442279590842669, "step": 4560}, {"loss": 1.2597, "grad_norm": 1.0147196054458618, "learning_rate": 0.0002, "epoch": 4.452021432050658, "step": 4570}, {"loss": 1.2148, "grad_norm": 0.9763361811637878, "learning_rate": 0.0002, "epoch": 4.461763273258645, "step": 4580}, {"loss": 1.3076, "grad_norm": 1.0300798416137695, "learning_rate": 0.0002, "epoch": 4.471505114466634, "step": 4590}, {"loss": 1.2665, "grad_norm": 0.8833121657371521, "learning_rate": 0.0002, "epoch": 4.481246955674623, "step": 4600}, {"loss": 1.1899, "grad_norm": 1.1214020252227783, "learning_rate": 0.0002, "epoch": 4.490988796882611, "step": 4610}, {"loss": 1.2579, "grad_norm": 0.8843787908554077, "learning_rate": 0.0002, "epoch": 4.500730638090599, "step": 4620}, {"loss": 1.2633, "grad_norm": 0.9942020773887634, "learning_rate": 0.0002, "epoch": 4.5104724792985875, "step": 4630}, {"loss": 1.3172, "grad_norm": 1.0033202171325684, "learning_rate": 0.0002, "epoch": 4.520214320506576, "step": 4640}, {"loss": 1.2024, "grad_norm": 0.8767235279083252, "learning_rate": 0.0002, "epoch": 4.529956161714564, "step": 4650}, {"loss": 1.2714, "grad_norm": 1.0117276906967163, "learning_rate": 0.0002, "epoch": 4.539698002922552, "step": 4660}, {"loss": 1.2911, "grad_norm": 1.2787362337112427, "learning_rate": 0.0002, "epoch": 4.549439844130541, "step": 4670}, {"loss": 1.2603, "grad_norm": 0.8824878931045532, "learning_rate": 0.0002, "epoch": 4.559181685338529, "step": 4680}, {"loss": 1.2905, "grad_norm": 0.9209560751914978, "learning_rate": 0.0002, "epoch": 4.568923526546517, "step": 4690}, {"loss": 1.1916, "grad_norm": 1.1064010858535767, "learning_rate": 0.0002, "epoch": 4.578665367754506, "step": 4700}, {"loss": 1.2217, "grad_norm": 0.8914572596549988, "learning_rate": 0.0002, "epoch": 4.588407208962494, "step": 4710}, {"loss": 1.2861, "grad_norm": 1.0412265062332153, "learning_rate": 0.0002, "epoch": 4.598149050170482, "step": 4720}, {"loss": 1.262, "grad_norm": 1.1950221061706543, "learning_rate": 0.0002, "epoch": 4.607890891378471, "step": 4730}, {"loss": 1.2659, "grad_norm": 0.8938062787055969, "learning_rate": 0.0002, "epoch": 4.617632732586459, "step": 4740}, {"loss": 1.2621, "grad_norm": 0.9849569201469421, "learning_rate": 0.0002, "epoch": 4.627374573794447, "step": 4750}, {"loss": 1.2341, "grad_norm": 1.0081515312194824, "learning_rate": 0.0002, "epoch": 4.637116415002436, "step": 4760}, {"loss": 1.2023, "grad_norm": 0.8566309213638306, "learning_rate": 0.0002, "epoch": 4.6468582562104235, "step": 4770}, {"loss": 1.2723, "grad_norm": 1.1750118732452393, "learning_rate": 0.0002, "epoch": 4.656600097418412, "step": 4780}, {"loss": 1.2537, "grad_norm": 0.925502598285675, "learning_rate": 0.0002, "epoch": 4.666341938626401, "step": 4790}, {"loss": 1.2146, "grad_norm": 1.0402472019195557, "learning_rate": 0.0002, "epoch": 4.676083779834388, "step": 4800}, {"loss": 1.2555, "grad_norm": 0.9772472977638245, "learning_rate": 0.0002, "epoch": 4.685825621042377, "step": 4810}, {"loss": 1.2667, "grad_norm": 0.9082779288291931, "learning_rate": 0.0002, "epoch": 4.695567462250366, "step": 4820}, {"loss": 1.2465, "grad_norm": 0.8026862740516663, "learning_rate": 0.0002, "epoch": 4.705309303458353, "step": 4830}, {"loss": 1.3369, "grad_norm": 1.1631089448928833, "learning_rate": 0.0002, "epoch": 4.715051144666342, "step": 4840}, {"loss": 1.261, "grad_norm": 0.9384787678718567, "learning_rate": 0.0002, "epoch": 4.7247929858743305, "step": 4850}, {"loss": 1.2588, "grad_norm": 1.2151581048965454, "learning_rate": 0.0002, "epoch": 4.734534827082318, "step": 4860}, {"loss": 1.363, "grad_norm": 0.9679436087608337, "learning_rate": 0.0002, "epoch": 4.744276668290307, "step": 4870}, {"loss": 1.3292, "grad_norm": 0.8352158069610596, "learning_rate": 0.0002, "epoch": 4.754018509498295, "step": 4880}, {"loss": 1.3056, "grad_norm": 1.0205804109573364, "learning_rate": 0.0002, "epoch": 4.763760350706283, "step": 4890}, {"loss": 1.223, "grad_norm": 0.9814772605895996, "learning_rate": 0.0002, "epoch": 4.773502191914272, "step": 4900}, {"loss": 1.3114, "grad_norm": 1.002854347229004, "learning_rate": 0.0002, "epoch": 4.78324403312226, "step": 4910}, {"loss": 1.3143, "grad_norm": 1.1609505414962769, "learning_rate": 0.0002, "epoch": 4.792985874330248, "step": 4920}, {"loss": 1.3166, "grad_norm": 0.9354982376098633, "learning_rate": 0.0002, "epoch": 4.802727715538237, "step": 4930}, {"loss": 1.2978, "grad_norm": 0.9761685729026794, "learning_rate": 0.0002, "epoch": 4.812469556746225, "step": 4940}, {"loss": 1.2709, "grad_norm": 1.0604596138000488, "learning_rate": 0.0002, "epoch": 4.822211397954213, "step": 4950}, {"loss": 1.2765, "grad_norm": 1.0902808904647827, "learning_rate": 0.0002, "epoch": 4.8319532391622015, "step": 4960}, {"loss": 1.3073, "grad_norm": 1.0174955129623413, "learning_rate": 0.0002, "epoch": 4.84169508037019, "step": 4970}, {"loss": 1.3141, "grad_norm": 1.0995253324508667, "learning_rate": 0.0002, "epoch": 4.851436921578179, "step": 4980}, {"loss": 1.3006, "grad_norm": 0.880993127822876, "learning_rate": 0.0002, "epoch": 4.8611787627861665, "step": 4990}, {"loss": 1.2547, "grad_norm": 0.9472237825393677, "learning_rate": 0.0002, "epoch": 4.870920603994155, "step": 5000}, {"loss": 1.4078, "grad_norm": 0.9504236578941345, "learning_rate": 0.0002, "epoch": 4.880662445202143, "step": 5010}, {"loss": 1.2791, "grad_norm": 1.1261742115020752, "learning_rate": 0.0002, "epoch": 4.890404286410131, "step": 5020}, {"loss": 1.3707, "grad_norm": 0.904674768447876, "learning_rate": 0.0002, "epoch": 4.90014612761812, "step": 5030}, {"loss": 1.2762, "grad_norm": 0.8828991055488586, "learning_rate": 0.0002, "epoch": 4.909887968826109, "step": 5040}, {"loss": 1.2905, "grad_norm": 1.0156532526016235, "learning_rate": 0.0002, "epoch": 4.919629810034096, "step": 5050}, {"loss": 1.3079, "grad_norm": 0.8975168466567993, "learning_rate": 0.0002, "epoch": 4.929371651242085, "step": 5060}, {"loss": 1.3322, "grad_norm": 0.9787213802337646, "learning_rate": 0.0002, "epoch": 4.939113492450073, "step": 5070}, {"loss": 1.2533, "grad_norm": 1.0801568031311035, "learning_rate": 0.0002, "epoch": 4.948855333658061, "step": 5080}, {"loss": 1.238, "grad_norm": 1.0655089616775513, "learning_rate": 0.0002, "epoch": 4.95859717486605, "step": 5090}, {"loss": 1.2449, "grad_norm": 0.8941320180892944, "learning_rate": 0.0002, "epoch": 4.968339016074038, "step": 5100}, {"loss": 1.2846, "grad_norm": 1.050621747970581, "learning_rate": 0.0002, "epoch": 4.978080857282026, "step": 5110}, {"loss": 1.3791, "grad_norm": 0.9724781513214111, "learning_rate": 0.0002, "epoch": 4.987822698490015, "step": 5120}, {"loss": 1.292, "grad_norm": 0.9850538969039917, "learning_rate": 0.0002, "epoch": 4.997564539698003, "step": 5130}, {"eval_loss": 2.0824170112609863, "eval_runtime": 55.592, "eval_samples_per_second": 9.12, "eval_steps_per_second": 1.151, "epoch": 4.9995129079396, "step": 5132}, {"loss": 1.037, "grad_norm": 1.0096189975738525, "learning_rate": 0.0002, "epoch": 5.007306380905991, "step": 5140}, {"loss": 1.0003, "grad_norm": 1.2403408288955688, "learning_rate": 0.0002, "epoch": 5.01704822211398, "step": 5150}, {"loss": 1.0129, "grad_norm": 1.1243221759796143, "learning_rate": 0.0002, "epoch": 5.026790063321968, "step": 5160}, {"loss": 0.9815, "grad_norm": 1.4745502471923828, "learning_rate": 0.0002, "epoch": 5.036531904529956, "step": 5170}, {"loss": 0.9715, "grad_norm": 1.1913198232650757, "learning_rate": 0.0002, "epoch": 5.0462737457379445, "step": 5180}, {"loss": 0.9282, "grad_norm": 1.2732855081558228, "learning_rate": 0.0002, "epoch": 5.056015586945933, "step": 5190}, {"loss": 0.9857, "grad_norm": 1.1737396717071533, "learning_rate": 0.0002, "epoch": 5.065757428153921, "step": 5200}, {"loss": 0.9754, "grad_norm": 1.4162768125534058, "learning_rate": 0.0002, "epoch": 5.075499269361909, "step": 5210}, {"loss": 1.0333, "grad_norm": 1.528274655342102, "learning_rate": 0.0002, "epoch": 5.085241110569898, "step": 5220}, {"loss": 1.0227, "grad_norm": 1.3966618776321411, "learning_rate": 0.0002, "epoch": 5.094982951777886, "step": 5230}, {"loss": 0.987, "grad_norm": 1.3427953720092773, "learning_rate": 0.0002, "epoch": 5.104724792985874, "step": 5240}, {"loss": 1.0353, "grad_norm": 1.6533905267715454, "learning_rate": 0.0002, "epoch": 5.114466634193863, "step": 5250}, {"loss": 1.0452, "grad_norm": 1.4114865064620972, "learning_rate": 0.0002, "epoch": 5.124208475401851, "step": 5260}, {"loss": 1.067, "grad_norm": 1.5460708141326904, "learning_rate": 0.0002, "epoch": 5.133950316609839, "step": 5270}, {"loss": 1.0667, "grad_norm": 1.3491919040679932, "learning_rate": 0.0002, "epoch": 5.143692157817828, "step": 5280}, {"loss": 0.9957, "grad_norm": 1.2208969593048096, "learning_rate": 0.0002, "epoch": 5.153433999025816, "step": 5290}, {"loss": 1.0362, "grad_norm": 1.1141403913497925, "learning_rate": 0.0002, "epoch": 5.163175840233804, "step": 5300}, {"loss": 0.9744, "grad_norm": 1.2938064336776733, "learning_rate": 0.0002, "epoch": 5.172917681441793, "step": 5310}, {"loss": 1.0438, "grad_norm": 1.2704918384552002, "learning_rate": 0.0002, "epoch": 5.1826595226497805, "step": 5320}, {"loss": 1.0015, "grad_norm": 1.3928544521331787, "learning_rate": 0.0002, "epoch": 5.192401363857769, "step": 5330}, {"loss": 1.025, "grad_norm": 1.1993824243545532, "learning_rate": 0.0002, "epoch": 5.202143205065758, "step": 5340}, {"loss": 1.0195, "grad_norm": 1.5913670063018799, "learning_rate": 0.0002, "epoch": 5.211885046273745, "step": 5350}, {"loss": 1.0113, "grad_norm": 1.1577855348587036, "learning_rate": 0.0002, "epoch": 5.221626887481734, "step": 5360}, {"loss": 1.0684, "grad_norm": 1.4535993337631226, "learning_rate": 0.0002, "epoch": 5.231368728689723, "step": 5370}, {"loss": 1.0255, "grad_norm": 1.5068976879119873, "learning_rate": 0.0002, "epoch": 5.24111056989771, "step": 5380}, {"loss": 1.0068, "grad_norm": 1.2365459203720093, "learning_rate": 0.0002, "epoch": 5.250852411105699, "step": 5390}, {"loss": 1.0145, "grad_norm": 1.3197922706604004, "learning_rate": 0.0002, "epoch": 5.2605942523136875, "step": 5400}, {"loss": 1.0767, "grad_norm": 1.2395117282867432, "learning_rate": 0.0002, "epoch": 5.270336093521675, "step": 5410}, {"loss": 1.0292, "grad_norm": 1.1841236352920532, "learning_rate": 0.0002, "epoch": 5.280077934729664, "step": 5420}, {"loss": 1.0233, "grad_norm": 1.218003749847412, "learning_rate": 0.0002, "epoch": 5.289819775937652, "step": 5430}, {"loss": 1.0093, "grad_norm": 1.2210947275161743, "learning_rate": 0.0002, "epoch": 5.29956161714564, "step": 5440}, {"loss": 0.9619, "grad_norm": 1.266006588935852, "learning_rate": 0.0002, "epoch": 5.309303458353629, "step": 5450}, {"loss": 1.0352, "grad_norm": 1.2598075866699219, "learning_rate": 0.0002, "epoch": 5.319045299561617, "step": 5460}, {"loss": 1.0929, "grad_norm": 1.2410019636154175, "learning_rate": 0.0002, "epoch": 5.328787140769606, "step": 5470}, {"loss": 1.058, "grad_norm": 1.249698519706726, "learning_rate": 0.0002, "epoch": 5.338528981977594, "step": 5480}, {"loss": 1.0457, "grad_norm": 1.2398173809051514, "learning_rate": 0.0002, "epoch": 5.348270823185582, "step": 5490}, {"loss": 1.0139, "grad_norm": 1.2416654825210571, "learning_rate": 0.0002, "epoch": 5.35801266439357, "step": 5500}, {"loss": 1.0609, "grad_norm": 1.398706316947937, "learning_rate": 0.0002, "epoch": 5.3677545056015585, "step": 5510}, {"loss": 1.0512, "grad_norm": 1.3049418926239014, "learning_rate": 0.0002, "epoch": 5.377496346809547, "step": 5520}, {"loss": 1.0912, "grad_norm": 1.2528893947601318, "learning_rate": 0.0002, "epoch": 5.387238188017536, "step": 5530}, {"loss": 1.0619, "grad_norm": 1.2963255643844604, "learning_rate": 0.0002, "epoch": 5.3969800292255234, "step": 5540}, {"loss": 1.0194, "grad_norm": 1.494231104850769, "learning_rate": 0.0002, "epoch": 5.406721870433512, "step": 5550}, {"loss": 1.0179, "grad_norm": 1.2760992050170898, "learning_rate": 0.0002, "epoch": 5.416463711641501, "step": 5560}, {"loss": 1.1088, "grad_norm": 1.195292592048645, "learning_rate": 0.0002, "epoch": 5.426205552849488, "step": 5570}, {"loss": 1.0859, "grad_norm": 1.6408965587615967, "learning_rate": 0.0002, "epoch": 5.435947394057477, "step": 5580}, {"loss": 1.0868, "grad_norm": 1.3092058897018433, "learning_rate": 0.0002, "epoch": 5.4456892352654656, "step": 5590}, {"loss": 1.006, "grad_norm": 1.2960586547851562, "learning_rate": 0.0002, "epoch": 5.455431076473453, "step": 5600}, {"loss": 1.0257, "grad_norm": 1.3560487031936646, "learning_rate": 0.0002, "epoch": 5.465172917681442, "step": 5610}, {"loss": 1.0314, "grad_norm": 1.1896311044692993, "learning_rate": 0.0002, "epoch": 5.4749147588894305, "step": 5620}, {"loss": 1.0435, "grad_norm": 1.3145595788955688, "learning_rate": 0.0002, "epoch": 5.484656600097418, "step": 5630}, {"loss": 1.0456, "grad_norm": 1.2207404375076294, "learning_rate": 0.0002, "epoch": 5.494398441305407, "step": 5640}, {"loss": 1.0823, "grad_norm": 1.266015887260437, "learning_rate": 0.0002, "epoch": 5.504140282513395, "step": 5650}, {"loss": 1.0696, "grad_norm": 1.2478289604187012, "learning_rate": 0.0002, "epoch": 5.513882123721383, "step": 5660}, {"loss": 1.0695, "grad_norm": 1.4851372241973877, "learning_rate": 0.0002, "epoch": 5.523623964929372, "step": 5670}, {"loss": 1.0736, "grad_norm": 1.4478679895401, "learning_rate": 0.0002, "epoch": 5.53336580613736, "step": 5680}, {"loss": 1.043, "grad_norm": 1.1079537868499756, "learning_rate": 0.0002, "epoch": 5.543107647345348, "step": 5690}, {"loss": 1.1107, "grad_norm": 1.4201879501342773, "learning_rate": 0.0002, "epoch": 5.552849488553337, "step": 5700}, {"loss": 1.0697, "grad_norm": 1.2092000246047974, "learning_rate": 0.0002, "epoch": 5.562591329761325, "step": 5710}, {"loss": 0.9868, "grad_norm": 1.4515851736068726, "learning_rate": 0.0002, "epoch": 5.572333170969313, "step": 5720}, {"loss": 1.1547, "grad_norm": 1.3260412216186523, "learning_rate": 0.0002, "epoch": 5.5820750121773015, "step": 5730}, {"loss": 1.1388, "grad_norm": 1.248191475868225, "learning_rate": 0.0002, "epoch": 5.59181685338529, "step": 5740}, {"loss": 1.0597, "grad_norm": 1.2037307024002075, "learning_rate": 0.0002, "epoch": 5.601558694593278, "step": 5750}, {"loss": 1.1425, "grad_norm": 1.341237187385559, "learning_rate": 0.0002, "epoch": 5.611300535801266, "step": 5760}, {"loss": 1.0942, "grad_norm": 1.130115270614624, "learning_rate": 0.0002, "epoch": 5.621042377009255, "step": 5770}, {"loss": 1.1029, "grad_norm": 1.3834772109985352, "learning_rate": 0.0002, "epoch": 5.630784218217243, "step": 5780}, {"loss": 1.0825, "grad_norm": 1.2586270570755005, "learning_rate": 0.0002, "epoch": 5.640526059425231, "step": 5790}, {"loss": 1.0186, "grad_norm": 1.3233023881912231, "learning_rate": 0.0002, "epoch": 5.65026790063322, "step": 5800}, {"loss": 1.0557, "grad_norm": 1.2711341381072998, "learning_rate": 0.0002, "epoch": 5.660009741841208, "step": 5810}, {"loss": 1.0897, "grad_norm": 1.3867720365524292, "learning_rate": 0.0002, "epoch": 5.669751583049196, "step": 5820}, {"loss": 1.0776, "grad_norm": 1.4783269166946411, "learning_rate": 0.0002, "epoch": 5.679493424257185, "step": 5830}, {"loss": 1.0632, "grad_norm": 1.2744768857955933, "learning_rate": 0.0002, "epoch": 5.6892352654651726, "step": 5840}, {"loss": 1.1484, "grad_norm": 1.3405882120132446, "learning_rate": 0.0002, "epoch": 5.698977106673161, "step": 5850}, {"loss": 1.0975, "grad_norm": 1.204300880432129, "learning_rate": 0.0002, "epoch": 5.70871894788115, "step": 5860}, {"loss": 1.0494, "grad_norm": 1.2954572439193726, "learning_rate": 0.0002, "epoch": 5.7184607890891375, "step": 5870}, {"loss": 1.0643, "grad_norm": 1.5478382110595703, "learning_rate": 0.0002, "epoch": 5.728202630297126, "step": 5880}, {"loss": 1.0582, "grad_norm": 1.2095842361450195, "learning_rate": 0.0002, "epoch": 5.737944471505115, "step": 5890}, {"loss": 1.1, "grad_norm": 1.0691519975662231, "learning_rate": 0.0002, "epoch": 5.747686312713103, "step": 5900}, {"loss": 1.0906, "grad_norm": 1.1920677423477173, "learning_rate": 0.0002, "epoch": 5.757428153921091, "step": 5910}, {"loss": 1.1746, "grad_norm": 1.2051277160644531, "learning_rate": 0.0002, "epoch": 5.76716999512908, "step": 5920}, {"loss": 1.1221, "grad_norm": 1.197490930557251, "learning_rate": 0.0002, "epoch": 5.776911836337067, "step": 5930}, {"loss": 1.07, "grad_norm": 1.2003998756408691, "learning_rate": 0.0002, "epoch": 5.786653677545056, "step": 5940}, {"loss": 1.0938, "grad_norm": 1.2323646545410156, "learning_rate": 0.0002, "epoch": 5.7963955187530445, "step": 5950}, {"loss": 1.1443, "grad_norm": 1.2593932151794434, "learning_rate": 0.0002, "epoch": 5.806137359961033, "step": 5960}, {"loss": 1.0829, "grad_norm": 1.1835976839065552, "learning_rate": 0.0002, "epoch": 5.815879201169021, "step": 5970}, {"loss": 1.1056, "grad_norm": 1.4770104885101318, "learning_rate": 0.0002, "epoch": 5.825621042377009, "step": 5980}, {"loss": 1.1934, "grad_norm": 1.1025809049606323, "learning_rate": 0.0002, "epoch": 5.835362883584997, "step": 5990}, {"loss": 1.1323, "grad_norm": 1.364588975906372, "learning_rate": 0.0002, "epoch": 5.845104724792986, "step": 6000}, {"loss": 1.1234, "grad_norm": 1.2340112924575806, "learning_rate": 0.0002, "epoch": 5.854846566000974, "step": 6010}, {"loss": 1.1123, "grad_norm": 1.4925711154937744, "learning_rate": 0.0002, "epoch": 5.864588407208963, "step": 6020}, {"loss": 1.12, "grad_norm": 1.3516744375228882, "learning_rate": 0.0002, "epoch": 5.874330248416951, "step": 6030}, {"loss": 1.1399, "grad_norm": 1.2058138847351074, "learning_rate": 0.0002, "epoch": 5.884072089624939, "step": 6040}, {"loss": 1.1074, "grad_norm": 1.13870108127594, "learning_rate": 0.0002, "epoch": 5.893813930832927, "step": 6050}, {"loss": 1.088, "grad_norm": 1.1587319374084473, "learning_rate": 0.0002, "epoch": 5.9035557720409155, "step": 6060}, {"loss": 1.1376, "grad_norm": 1.164481520652771, "learning_rate": 0.0002, "epoch": 5.913297613248904, "step": 6070}, {"loss": 1.1262, "grad_norm": 1.2115206718444824, "learning_rate": 0.0002, "epoch": 5.923039454456893, "step": 6080}, {"loss": 1.1345, "grad_norm": 1.3201590776443481, "learning_rate": 0.0002, "epoch": 5.93278129566488, "step": 6090}, {"loss": 1.1288, "grad_norm": 1.287380576133728, "learning_rate": 0.0002, "epoch": 5.942523136872869, "step": 6100}, {"loss": 1.1475, "grad_norm": 1.1820166110992432, "learning_rate": 0.0002, "epoch": 5.952264978080858, "step": 6110}, {"loss": 1.1112, "grad_norm": 1.2550667524337769, "learning_rate": 0.0002, "epoch": 5.962006819288845, "step": 6120}, {"loss": 1.1528, "grad_norm": 1.3547813892364502, "learning_rate": 0.0002, "epoch": 5.971748660496834, "step": 6130}, {"loss": 1.0557, "grad_norm": 1.260842204093933, "learning_rate": 0.0002, "epoch": 5.9814905017048225, "step": 6140}, {"loss": 1.1119, "grad_norm": 1.1643036603927612, "learning_rate": 0.0002, "epoch": 5.99123234291281, "step": 6150}, {"eval_loss": 2.2628161907196045, "eval_runtime": 57.2379, "eval_samples_per_second": 8.858, "eval_steps_per_second": 1.118, "epoch": 6.0, "step": 6159}, {"loss": 1.0837, "grad_norm": 0.9384723901748657, "learning_rate": 0.0002, "epoch": 6.000974184120799, "step": 6160}, {"loss": 0.7335, "grad_norm": 2.1525821685791016, "learning_rate": 0.0002, "epoch": 6.0107160253287875, "step": 6170}, {"loss": 0.8416, "grad_norm": 2.0194077491760254, "learning_rate": 0.0002, "epoch": 6.020457866536775, "step": 6180}, {"loss": 0.8443, "grad_norm": 1.5257816314697266, "learning_rate": 0.0002, "epoch": 6.030199707744764, "step": 6190}, {"loss": 0.7543, "grad_norm": 1.5432662963867188, "learning_rate": 0.0002, "epoch": 6.039941548952752, "step": 6200}, {"loss": 0.8104, "grad_norm": 1.6874405145645142, "learning_rate": 0.0002, "epoch": 6.04968339016074, "step": 6210}, {"loss": 0.8395, "grad_norm": 1.7346407175064087, "learning_rate": 0.0002, "epoch": 6.059425231368729, "step": 6220}, {"loss": 0.8027, "grad_norm": 1.5320781469345093, "learning_rate": 0.0002, "epoch": 6.069167072576717, "step": 6230}, {"loss": 0.7488, "grad_norm": 1.4106669425964355, "learning_rate": 0.0002, "epoch": 6.078908913784705, "step": 6240}, {"loss": 0.812, "grad_norm": 1.5568628311157227, "learning_rate": 0.0002, "epoch": 6.088650754992694, "step": 6250}, {"loss": 0.8055, "grad_norm": 1.6155978441238403, "learning_rate": 0.0002, "epoch": 6.098392596200682, "step": 6260}, {"loss": 0.8225, "grad_norm": 1.4820445775985718, "learning_rate": 0.0002, "epoch": 6.10813443740867, "step": 6270}, {"loss": 0.8599, "grad_norm": 1.6163820028305054, "learning_rate": 0.0002, "epoch": 6.1178762786166585, "step": 6280}, {"loss": 0.853, "grad_norm": 1.8396387100219727, "learning_rate": 0.0002, "epoch": 6.127618119824647, "step": 6290}, {"loss": 0.7768, "grad_norm": 1.7181230783462524, "learning_rate": 0.0002, "epoch": 6.137359961032635, "step": 6300}, {"loss": 0.8116, "grad_norm": 1.6568509340286255, "learning_rate": 0.0002, "epoch": 6.147101802240623, "step": 6310}, {"loss": 0.8525, "grad_norm": 1.3481947183609009, "learning_rate": 0.0002, "epoch": 6.156843643448612, "step": 6320}, {"loss": 0.762, "grad_norm": 1.5788342952728271, "learning_rate": 0.0002, "epoch": 6.1665854846566, "step": 6330}, {"loss": 0.886, "grad_norm": 1.5067620277404785, "learning_rate": 0.0002, "epoch": 6.176327325864588, "step": 6340}, {"loss": 0.8375, "grad_norm": 1.8198208808898926, "learning_rate": 0.0002, "epoch": 6.186069167072577, "step": 6350}, {"loss": 0.7867, "grad_norm": 1.4012749195098877, "learning_rate": 0.0002, "epoch": 6.195811008280565, "step": 6360}, {"loss": 0.8144, "grad_norm": 1.759798288345337, "learning_rate": 0.0002, "epoch": 6.205552849488553, "step": 6370}, {"loss": 0.7811, "grad_norm": 1.468922734260559, "learning_rate": 0.0002, "epoch": 6.215294690696542, "step": 6380}, {"loss": 0.8356, "grad_norm": 1.3706471920013428, "learning_rate": 0.0002, "epoch": 6.2250365319045295, "step": 6390}, {"loss": 0.8096, "grad_norm": 1.6397383213043213, "learning_rate": 0.0002, "epoch": 6.234778373112518, "step": 6400}, {"loss": 0.8834, "grad_norm": 1.5614187717437744, "learning_rate": 0.0002, "epoch": 6.244520214320507, "step": 6410}, {"loss": 0.8533, "grad_norm": 1.7118678092956543, "learning_rate": 0.0002, "epoch": 6.2542620555284945, "step": 6420}, {"loss": 0.8653, "grad_norm": 1.4041547775268555, "learning_rate": 0.0002, "epoch": 6.264003896736483, "step": 6430}, {"loss": 0.879, "grad_norm": 1.7653605937957764, "learning_rate": 0.0002, "epoch": 6.273745737944472, "step": 6440}, {"loss": 0.8786, "grad_norm": 2.6219191551208496, "learning_rate": 0.0002, "epoch": 6.28348757915246, "step": 6450}, {"loss": 0.8896, "grad_norm": 1.4757837057113647, "learning_rate": 0.0002, "epoch": 6.293229420360448, "step": 6460}, {"loss": 0.9079, "grad_norm": 1.715598225593567, "learning_rate": 0.0002, "epoch": 6.302971261568437, "step": 6470}, {"loss": 0.8526, "grad_norm": 1.376216173171997, "learning_rate": 0.0002, "epoch": 6.312713102776424, "step": 6480}, {"loss": 0.8742, "grad_norm": 1.7119828462600708, "learning_rate": 0.0002, "epoch": 6.322454943984413, "step": 6490}, {"loss": 0.7988, "grad_norm": 1.4304355382919312, "learning_rate": 0.0002, "epoch": 6.3321967851924015, "step": 6500}, {"loss": 0.8539, "grad_norm": 1.4889872074127197, "learning_rate": 0.0002, "epoch": 6.34193862640039, "step": 6510}, {"loss": 0.9328, "grad_norm": 1.370373010635376, "learning_rate": 0.0002, "epoch": 6.351680467608378, "step": 6520}, {"loss": 0.8997, "grad_norm": 1.7697709798812866, "learning_rate": 0.0002, "epoch": 6.361422308816366, "step": 6530}, {"loss": 0.9421, "grad_norm": 1.495297908782959, "learning_rate": 0.0002, "epoch": 6.371164150024355, "step": 6540}, {"loss": 0.8796, "grad_norm": 1.7251347303390503, "learning_rate": 0.0002, "epoch": 6.380905991232343, "step": 6550}, {"loss": 0.9327, "grad_norm": 1.6909505128860474, "learning_rate": 0.0002, "epoch": 6.390647832440331, "step": 6560}, {"loss": 0.837, "grad_norm": 1.4369314908981323, "learning_rate": 0.0002, "epoch": 6.40038967364832, "step": 6570}, {"loss": 0.8572, "grad_norm": 1.7803739309310913, "learning_rate": 0.0002, "epoch": 6.410131514856308, "step": 6580}, {"loss": 0.9024, "grad_norm": 1.6107097864151, "learning_rate": 0.0002, "epoch": 6.419873356064296, "step": 6590}, {"loss": 0.8469, "grad_norm": 1.6151643991470337, "learning_rate": 0.0002, "epoch": 6.429615197272285, "step": 6600}, {"loss": 0.8791, "grad_norm": 1.7159833908081055, "learning_rate": 0.0002, "epoch": 6.4393570384802725, "step": 6610}, {"loss": 0.9249, "grad_norm": 1.4366064071655273, "learning_rate": 0.0002, "epoch": 6.449098879688261, "step": 6620}, {"loss": 0.8417, "grad_norm": 1.6050453186035156, "learning_rate": 0.0002, "epoch": 6.45884072089625, "step": 6630}, {"loss": 0.8943, "grad_norm": 1.6296740770339966, "learning_rate": 0.0002, "epoch": 6.468582562104237, "step": 6640}, {"loss": 0.9228, "grad_norm": 1.6181174516677856, "learning_rate": 0.0002, "epoch": 6.478324403312226, "step": 6650}, {"loss": 0.9139, "grad_norm": 1.5452176332473755, "learning_rate": 0.0002, "epoch": 6.488066244520215, "step": 6660}, {"loss": 0.9022, "grad_norm": 1.3919731378555298, "learning_rate": 0.0002, "epoch": 6.497808085728202, "step": 6670}, {"loss": 0.9046, "grad_norm": 1.6456257104873657, "learning_rate": 0.0002, "epoch": 6.507549926936191, "step": 6680}, {"loss": 0.9041, "grad_norm": 1.4147369861602783, "learning_rate": 0.0002, "epoch": 6.5172917681441795, "step": 6690}, {"loss": 0.8361, "grad_norm": 1.7005025148391724, "learning_rate": 0.0002, "epoch": 6.527033609352167, "step": 6700}, {"loss": 0.8738, "grad_norm": 1.6032357215881348, "learning_rate": 0.0002, "epoch": 6.536775450560156, "step": 6710}, {"loss": 0.9796, "grad_norm": 1.3454229831695557, "learning_rate": 0.0002, "epoch": 6.5465172917681445, "step": 6720}, {"loss": 0.8573, "grad_norm": 1.6961418390274048, "learning_rate": 0.0002, "epoch": 6.556259132976132, "step": 6730}, {"loss": 0.9241, "grad_norm": 1.78407883644104, "learning_rate": 0.0002, "epoch": 6.566000974184121, "step": 6740}, {"loss": 0.8941, "grad_norm": 1.6817889213562012, "learning_rate": 0.0002, "epoch": 6.575742815392109, "step": 6750}, {"loss": 0.8765, "grad_norm": 1.7894943952560425, "learning_rate": 0.0002, "epoch": 6.585484656600097, "step": 6760}, {"loss": 0.8607, "grad_norm": 1.6404837369918823, "learning_rate": 0.0002, "epoch": 6.595226497808086, "step": 6770}, {"loss": 0.8573, "grad_norm": 1.5849255323410034, "learning_rate": 0.0002, "epoch": 6.604968339016074, "step": 6780}, {"loss": 0.9575, "grad_norm": 1.5993813276290894, "learning_rate": 0.0002, "epoch": 6.614710180224062, "step": 6790}, {"loss": 0.8922, "grad_norm": 1.2834863662719727, "learning_rate": 0.0002, "epoch": 6.624452021432051, "step": 6800}, {"loss": 0.9007, "grad_norm": 1.7215641736984253, "learning_rate": 0.0002, "epoch": 6.634193862640039, "step": 6810}, {"loss": 0.9292, "grad_norm": 1.7588146924972534, "learning_rate": 0.0002, "epoch": 6.643935703848027, "step": 6820}, {"loss": 0.8634, "grad_norm": 1.7956023216247559, "learning_rate": 0.0002, "epoch": 6.6536775450560155, "step": 6830}, {"loss": 0.8108, "grad_norm": 1.5115351676940918, "learning_rate": 0.0002, "epoch": 6.663419386264004, "step": 6840}, {"loss": 0.9329, "grad_norm": 1.5660319328308105, "learning_rate": 0.0002, "epoch": 6.673161227471992, "step": 6850}, {"loss": 0.9877, "grad_norm": 1.4323679208755493, "learning_rate": 0.0002, "epoch": 6.68290306867998, "step": 6860}, {"loss": 0.8732, "grad_norm": 1.662089467048645, "learning_rate": 0.0002, "epoch": 6.692644909887969, "step": 6870}, {"loss": 0.87, "grad_norm": 1.7854869365692139, "learning_rate": 0.0002, "epoch": 6.702386751095958, "step": 6880}, {"loss": 0.9105, "grad_norm": 1.5491222143173218, "learning_rate": 0.0002, "epoch": 6.712128592303945, "step": 6890}, {"loss": 0.9147, "grad_norm": 1.5946987867355347, "learning_rate": 0.0002, "epoch": 6.721870433511934, "step": 6900}, {"loss": 0.9391, "grad_norm": 1.6195964813232422, "learning_rate": 0.0002, "epoch": 6.731612274719922, "step": 6910}, {"loss": 0.8947, "grad_norm": 1.6366901397705078, "learning_rate": 0.0002, "epoch": 6.74135411592791, "step": 6920}, {"loss": 0.8695, "grad_norm": 1.5080382823944092, "learning_rate": 0.0002, "epoch": 6.751095957135899, "step": 6930}, {"loss": 0.9124, "grad_norm": 1.742353916168213, "learning_rate": 0.0002, "epoch": 6.760837798343887, "step": 6940}, {"loss": 0.9118, "grad_norm": 1.690251111984253, "learning_rate": 0.0002, "epoch": 6.770579639551875, "step": 6950}, {"loss": 0.9039, "grad_norm": 1.7103357315063477, "learning_rate": 0.0002, "epoch": 6.780321480759864, "step": 6960}, {"loss": 0.869, "grad_norm": 1.6630914211273193, "learning_rate": 0.0002, "epoch": 6.7900633219678515, "step": 6970}, {"loss": 0.8944, "grad_norm": 1.423768162727356, "learning_rate": 0.0002, "epoch": 6.79980516317584, "step": 6980}, {"loss": 0.9397, "grad_norm": 1.7844693660736084, "learning_rate": 0.0002, "epoch": 6.809547004383829, "step": 6990}, {"loss": 0.8889, "grad_norm": 1.545282006263733, "learning_rate": 0.0002, "epoch": 6.819288845591817, "step": 7000}, {"loss": 0.9333, "grad_norm": 1.4340319633483887, "learning_rate": 0.0002, "epoch": 6.829030686799805, "step": 7010}, {"loss": 0.9486, "grad_norm": 1.5981626510620117, "learning_rate": 0.0002, "epoch": 6.838772528007794, "step": 7020}, {"loss": 0.9062, "grad_norm": 1.5205026865005493, "learning_rate": 0.0002, "epoch": 6.848514369215782, "step": 7030}, {"loss": 0.9245, "grad_norm": 1.6999989748001099, "learning_rate": 0.0002, "epoch": 6.85825621042377, "step": 7040}, {"loss": 0.9313, "grad_norm": 1.6392347812652588, "learning_rate": 0.0002, "epoch": 6.8679980516317585, "step": 7050}, {"loss": 0.9275, "grad_norm": 1.637308955192566, "learning_rate": 0.0002, "epoch": 6.877739892839747, "step": 7060}, {"loss": 0.9672, "grad_norm": 1.671341896057129, "learning_rate": 0.0002, "epoch": 6.887481734047735, "step": 7070}, {"loss": 0.9726, "grad_norm": 1.4437555074691772, "learning_rate": 0.0002, "epoch": 6.897223575255723, "step": 7080}, {"loss": 0.9454, "grad_norm": 1.4251935482025146, "learning_rate": 0.0002, "epoch": 6.906965416463712, "step": 7090}, {"loss": 0.8858, "grad_norm": 1.5106734037399292, "learning_rate": 0.0002, "epoch": 6.9167072576717, "step": 7100}, {"loss": 0.939, "grad_norm": 1.670742154121399, "learning_rate": 0.0002, "epoch": 6.926449098879688, "step": 7110}, {"loss": 0.8818, "grad_norm": 1.4353723526000977, "learning_rate": 0.0002, "epoch": 6.936190940087677, "step": 7120}, {"loss": 0.9354, "grad_norm": 1.9437772035598755, "learning_rate": 0.0002, "epoch": 6.945932781295665, "step": 7130}, {"loss": 0.9623, "grad_norm": 1.4922038316726685, "learning_rate": 0.0002, "epoch": 6.955674622503653, "step": 7140}, {"loss": 0.9653, "grad_norm": 1.489193081855774, "learning_rate": 0.0002, "epoch": 6.965416463711642, "step": 7150}, {"loss": 1.0024, "grad_norm": 1.529490351676941, "learning_rate": 0.0002, "epoch": 6.9751583049196295, "step": 7160}, {"loss": 0.9715, "grad_norm": 1.7370105981826782, "learning_rate": 0.0002, "epoch": 6.984900146127618, "step": 7170}, {"loss": 0.921, "grad_norm": 1.5639604330062866, "learning_rate": 0.0002, "epoch": 6.994641987335607, "step": 7180}]}
+{"epoch": 7.996103263516805, "step": 8208, "epoch_duration": 1626.2399775981903, "total_accumulated_duration": 14136.657781124115, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-4/checkpoint-1026", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 3.0782, "grad_norm": 0.6537588834762573, "learning_rate": 0.0002, "epoch": 0.00974184120798831, "step": 10}, {"loss": 2.4644, "grad_norm": 0.5270306468009949, "learning_rate": 0.0002, "epoch": 0.01948368241597662, "step": 20}, {"loss": 2.2589, "grad_norm": 0.6826501488685608, "learning_rate": 0.0002, "epoch": 0.029225523623964928, "step": 30}, {"loss": 2.0141, "grad_norm": 0.5061377286911011, "learning_rate": 0.0002, "epoch": 0.03896736483195324, "step": 40}, {"loss": 1.9458, "grad_norm": 0.4300410747528076, "learning_rate": 0.0002, "epoch": 0.04870920603994155, "step": 50}, {"loss": 1.983, "grad_norm": 0.5063319802284241, "learning_rate": 0.0002, "epoch": 0.058451047247929856, "step": 60}, {"loss": 1.9799, "grad_norm": 0.49310117959976196, "learning_rate": 0.0002, "epoch": 0.06819288845591817, "step": 70}, {"loss": 1.9277, "grad_norm": 0.4676004648208618, "learning_rate": 0.0002, "epoch": 0.07793472966390648, "step": 80}, {"loss": 1.9147, "grad_norm": 0.41647228598594666, "learning_rate": 0.0002, "epoch": 0.08767657087189479, "step": 90}, {"loss": 1.8894, "grad_norm": 0.40217313170433044, "learning_rate": 0.0002, "epoch": 0.0974184120798831, "step": 100}, {"loss": 1.9099, "grad_norm": 0.4123637080192566, "learning_rate": 0.0002, "epoch": 0.1071602532878714, "step": 110}, {"loss": 1.8471, "grad_norm": 0.37414297461509705, "learning_rate": 0.0002, "epoch": 0.11690209449585971, "step": 120}, {"loss": 1.894, "grad_norm": 0.3179326355457306, "learning_rate": 0.0002, "epoch": 0.12664393570384802, "step": 130}, {"loss": 1.847, "grad_norm": 0.3548192083835602, "learning_rate": 0.0002, "epoch": 0.13638577691183634, "step": 140}, {"loss": 1.7919, "grad_norm": 0.3273540139198303, "learning_rate": 0.0002, "epoch": 0.14612761811982464, "step": 150}, {"loss": 1.8496, "grad_norm": 0.36500975489616394, "learning_rate": 0.0002, "epoch": 0.15586945932781296, "step": 160}, {"loss": 1.8473, "grad_norm": 0.4106619656085968, "learning_rate": 0.0002, "epoch": 0.16561130053580125, "step": 170}, {"loss": 1.841, "grad_norm": 0.41361644864082336, "learning_rate": 0.0002, "epoch": 0.17535314174378958, "step": 180}, {"loss": 1.8879, "grad_norm": 0.3608580231666565, "learning_rate": 0.0002, "epoch": 0.1850949829517779, "step": 190}, {"loss": 1.7717, "grad_norm": 0.4291760325431824, "learning_rate": 0.0002, "epoch": 0.1948368241597662, "step": 200}, {"loss": 1.8437, "grad_norm": 0.344184011220932, "learning_rate": 0.0002, "epoch": 0.20457866536775451, "step": 210}, {"loss": 1.8779, "grad_norm": 0.3834705650806427, "learning_rate": 0.0002, "epoch": 0.2143205065757428, "step": 220}, {"loss": 1.7533, "grad_norm": 0.3738210201263428, "learning_rate": 0.0002, "epoch": 0.22406234778373113, "step": 230}, {"loss": 1.824, "grad_norm": 0.4306780695915222, "learning_rate": 0.0002, "epoch": 0.23380418899171942, "step": 240}, {"loss": 1.8519, "grad_norm": 0.5066465139389038, "learning_rate": 0.0002, "epoch": 0.24354603019970775, "step": 250}, {"loss": 1.7402, "grad_norm": 0.34227681159973145, "learning_rate": 0.0002, "epoch": 0.25328787140769604, "step": 260}, {"loss": 1.8614, "grad_norm": 0.3346865475177765, "learning_rate": 0.0002, "epoch": 0.26302971261568436, "step": 270}, {"loss": 1.8502, "grad_norm": 0.3639362156391144, "learning_rate": 0.0002, "epoch": 0.2727715538236727, "step": 280}, {"loss": 1.8428, "grad_norm": 0.33223700523376465, "learning_rate": 0.0002, "epoch": 0.282513395031661, "step": 290}, {"loss": 1.821, "grad_norm": 0.35176315903663635, "learning_rate": 0.0002, "epoch": 0.2922552362396493, "step": 300}, {"loss": 1.7635, "grad_norm": 0.3581472635269165, "learning_rate": 0.0002, "epoch": 0.3019970774476376, "step": 310}, {"loss": 1.8262, "grad_norm": 0.35943421721458435, "learning_rate": 0.0002, "epoch": 0.3117389186556259, "step": 320}, {"loss": 1.8167, "grad_norm": 0.322051078081131, "learning_rate": 0.0002, "epoch": 0.32148075986361424, "step": 330}, {"loss": 1.8221, "grad_norm": 0.33904823660850525, "learning_rate": 0.0002, "epoch": 0.3312226010716025, "step": 340}, {"loss": 1.9159, "grad_norm": 0.39162731170654297, "learning_rate": 0.0002, "epoch": 0.34096444227959083, "step": 350}, {"loss": 1.7966, "grad_norm": 0.330624520778656, "learning_rate": 0.0002, "epoch": 0.35070628348757915, "step": 360}, {"loss": 1.8646, "grad_norm": 0.3793248236179352, "learning_rate": 0.0002, "epoch": 0.3604481246955675, "step": 370}, {"loss": 1.8038, "grad_norm": 0.3347395658493042, "learning_rate": 0.0002, "epoch": 0.3701899659035558, "step": 380}, {"loss": 1.9244, "grad_norm": 0.30527254939079285, "learning_rate": 0.0002, "epoch": 0.37993180711154406, "step": 390}, {"loss": 1.7982, "grad_norm": 0.3081390857696533, "learning_rate": 0.0002, "epoch": 0.3896736483195324, "step": 400}, {"loss": 1.8968, "grad_norm": 0.3742620050907135, "learning_rate": 0.0002, "epoch": 0.3994154895275207, "step": 410}, {"loss": 1.8095, "grad_norm": 0.4080568253993988, "learning_rate": 0.0002, "epoch": 0.40915733073550903, "step": 420}, {"loss": 1.8555, "grad_norm": 0.38034746050834656, "learning_rate": 0.0002, "epoch": 0.4188991719434973, "step": 430}, {"loss": 1.8494, "grad_norm": 0.34893402457237244, "learning_rate": 0.0002, "epoch": 0.4286410131514856, "step": 440}, {"loss": 1.8481, "grad_norm": 0.33285608887672424, "learning_rate": 0.0002, "epoch": 0.43838285435947394, "step": 450}, {"loss": 1.8466, "grad_norm": 0.4110095798969269, "learning_rate": 0.0002, "epoch": 0.44812469556746226, "step": 460}, {"loss": 1.7906, "grad_norm": 0.3658817410469055, "learning_rate": 0.0002, "epoch": 0.4578665367754506, "step": 470}, {"loss": 1.7589, "grad_norm": 0.31350770592689514, "learning_rate": 0.0002, "epoch": 0.46760837798343885, "step": 480}, {"loss": 1.7839, "grad_norm": 0.38827991485595703, "learning_rate": 0.0002, "epoch": 0.47735021919142717, "step": 490}, {"loss": 1.8224, "grad_norm": 0.3792393207550049, "learning_rate": 0.0002, "epoch": 0.4870920603994155, "step": 500}, {"loss": 1.8028, "grad_norm": 0.3004095256328583, "learning_rate": 0.0002, "epoch": 0.4968339016074038, "step": 510}, {"loss": 1.6899, "grad_norm": 0.3200063407421112, "learning_rate": 0.0002, "epoch": 0.5065757428153921, "step": 520}, {"loss": 1.8205, "grad_norm": 0.3206128478050232, "learning_rate": 0.0002, "epoch": 0.5163175840233805, "step": 530}, {"loss": 1.7725, "grad_norm": 0.30258631706237793, "learning_rate": 0.0002, "epoch": 0.5260594252313687, "step": 540}, {"loss": 1.7791, "grad_norm": 0.28210392594337463, "learning_rate": 0.0002, "epoch": 0.535801266439357, "step": 550}, {"loss": 1.8221, "grad_norm": 0.34854066371917725, "learning_rate": 0.0002, "epoch": 0.5455431076473454, "step": 560}, {"loss": 1.8331, "grad_norm": 0.31689873337745667, "learning_rate": 0.0002, "epoch": 0.5552849488553336, "step": 570}, {"loss": 1.8311, "grad_norm": 0.31253790855407715, "learning_rate": 0.0002, "epoch": 0.565026790063322, "step": 580}, {"loss": 1.7035, "grad_norm": 0.3229721188545227, "learning_rate": 0.0002, "epoch": 0.5747686312713103, "step": 590}, {"loss": 1.766, "grad_norm": 0.3723772466182709, "learning_rate": 0.0002, "epoch": 0.5845104724792985, "step": 600}, {"loss": 1.8357, "grad_norm": 0.345798522233963, "learning_rate": 0.0002, "epoch": 0.5942523136872869, "step": 610}, {"loss": 1.7266, "grad_norm": 0.3440598249435425, "learning_rate": 0.0002, "epoch": 0.6039941548952752, "step": 620}, {"loss": 1.8307, "grad_norm": 0.3406416177749634, "learning_rate": 0.0002, "epoch": 0.6137359961032636, "step": 630}, {"loss": 1.8103, "grad_norm": 0.3218357264995575, "learning_rate": 0.0002, "epoch": 0.6234778373112518, "step": 640}, {"loss": 1.7457, "grad_norm": 0.45319172739982605, "learning_rate": 0.0002, "epoch": 0.6332196785192401, "step": 650}, {"loss": 1.8674, "grad_norm": 0.2787110507488251, "learning_rate": 0.0002, "epoch": 0.6429615197272285, "step": 660}, {"loss": 1.8426, "grad_norm": 0.3064707815647125, "learning_rate": 0.0002, "epoch": 0.6527033609352167, "step": 670}, {"loss": 1.846, "grad_norm": 0.2940629720687866, "learning_rate": 0.0002, "epoch": 0.662445202143205, "step": 680}, {"loss": 1.7865, "grad_norm": 0.31695225834846497, "learning_rate": 0.0002, "epoch": 0.6721870433511934, "step": 690}, {"loss": 1.8449, "grad_norm": 0.29589611291885376, "learning_rate": 0.0002, "epoch": 0.6819288845591817, "step": 700}, {"loss": 1.7828, "grad_norm": 0.3062121570110321, "learning_rate": 0.0002, "epoch": 0.69167072576717, "step": 710}, {"loss": 1.8108, "grad_norm": 0.3315656781196594, "learning_rate": 0.0002, "epoch": 0.7014125669751583, "step": 720}, {"loss": 1.8056, "grad_norm": 0.30353930592536926, "learning_rate": 0.0002, "epoch": 0.7111544081831466, "step": 730}, {"loss": 1.7682, "grad_norm": 0.28360483050346375, "learning_rate": 0.0002, "epoch": 0.720896249391135, "step": 740}, {"loss": 1.7386, "grad_norm": 0.3362562656402588, "learning_rate": 0.0002, "epoch": 0.7306380905991232, "step": 750}, {"loss": 1.7607, "grad_norm": 0.40434667468070984, "learning_rate": 0.0002, "epoch": 0.7403799318071116, "step": 760}, {"loss": 1.8374, "grad_norm": 0.2930425703525543, "learning_rate": 0.0002, "epoch": 0.7501217730150999, "step": 770}, {"loss": 1.8216, "grad_norm": 0.30177003145217896, "learning_rate": 0.0002, "epoch": 0.7598636142230881, "step": 780}, {"loss": 1.8081, "grad_norm": 0.2784474790096283, "learning_rate": 0.0002, "epoch": 0.7696054554310765, "step": 790}, {"loss": 1.7953, "grad_norm": 0.35849854350090027, "learning_rate": 0.0002, "epoch": 0.7793472966390648, "step": 800}, {"loss": 1.7896, "grad_norm": 0.27329114079475403, "learning_rate": 0.0002, "epoch": 0.7890891378470531, "step": 810}, {"loss": 1.7633, "grad_norm": 0.33331671357154846, "learning_rate": 0.0002, "epoch": 0.7988309790550414, "step": 820}, {"loss": 1.7508, "grad_norm": 0.28727295994758606, "learning_rate": 0.0002, "epoch": 0.8085728202630297, "step": 830}, {"loss": 1.7541, "grad_norm": 0.31391268968582153, "learning_rate": 0.0002, "epoch": 0.8183146614710181, "step": 840}, {"loss": 1.793, "grad_norm": 0.3303709030151367, "learning_rate": 0.0002, "epoch": 0.8280565026790063, "step": 850}, {"loss": 1.7479, "grad_norm": 0.33772537112236023, "learning_rate": 0.0002, "epoch": 0.8377983438869946, "step": 860}, {"loss": 1.7138, "grad_norm": 0.32876333594322205, "learning_rate": 0.0002, "epoch": 0.847540185094983, "step": 870}, {"loss": 1.7832, "grad_norm": 0.28444716334342957, "learning_rate": 0.0002, "epoch": 0.8572820263029712, "step": 880}, {"loss": 1.7077, "grad_norm": 0.3070019483566284, "learning_rate": 0.0002, "epoch": 0.8670238675109596, "step": 890}, {"loss": 1.7699, "grad_norm": 0.29484760761260986, "learning_rate": 0.0002, "epoch": 0.8767657087189479, "step": 900}, {"loss": 1.7211, "grad_norm": 0.32373034954071045, "learning_rate": 0.0002, "epoch": 0.8865075499269361, "step": 910}, {"loss": 1.7799, "grad_norm": 0.3229396939277649, "learning_rate": 0.0002, "epoch": 0.8962493911349245, "step": 920}, {"loss": 1.8226, "grad_norm": 0.33151453733444214, "learning_rate": 0.0002, "epoch": 0.9059912323429128, "step": 930}, {"loss": 1.8339, "grad_norm": 0.32037460803985596, "learning_rate": 0.0002, "epoch": 0.9157330735509012, "step": 940}, {"loss": 1.822, "grad_norm": 0.31283533573150635, "learning_rate": 0.0002, "epoch": 0.9254749147588894, "step": 950}, {"loss": 1.8233, "grad_norm": 0.27984118461608887, "learning_rate": 0.0002, "epoch": 0.9352167559668777, "step": 960}, {"loss": 1.7755, "grad_norm": 0.316500186920166, "learning_rate": 0.0002, "epoch": 0.9449585971748661, "step": 970}, {"loss": 1.8032, "grad_norm": 0.33708682656288147, "learning_rate": 0.0002, "epoch": 0.9547004383828543, "step": 980}, {"loss": 1.8863, "grad_norm": 0.31026017665863037, "learning_rate": 0.0002, "epoch": 0.9644422795908427, "step": 990}, {"loss": 1.8458, "grad_norm": 0.30874672532081604, "learning_rate": 0.0002, "epoch": 0.974184120798831, "step": 1000}, {"loss": 1.7975, "grad_norm": 0.3257741630077362, "learning_rate": 0.0002, "epoch": 0.9839259620068193, "step": 1010}, {"loss": 1.7936, "grad_norm": 0.2865653932094574, "learning_rate": 0.0002, "epoch": 0.9936678032148076, "step": 1020}, {"eval_loss": 1.8103164434432983, "eval_runtime": 56.3917, "eval_samples_per_second": 8.991, "eval_steps_per_second": 1.135, "epoch": 0.9995129079396006, "step": 1026}, {"loss": 1.7013, "grad_norm": 0.2860608398914337, "learning_rate": 0.0002, "epoch": 1.003409644422796, "step": 1030}, {"loss": 1.7521, "grad_norm": 0.3156210780143738, "learning_rate": 0.0002, "epoch": 1.0131514856307842, "step": 1040}, {"loss": 1.6182, "grad_norm": 0.26126575469970703, "learning_rate": 0.0002, "epoch": 1.0228933268387725, "step": 1050}, {"loss": 1.7546, "grad_norm": 0.3019633889198303, "learning_rate": 0.0002, "epoch": 1.032635168046761, "step": 1060}, {"loss": 1.7096, "grad_norm": 0.534140944480896, "learning_rate": 0.0002, "epoch": 1.042377009254749, "step": 1070}, {"loss": 1.664, "grad_norm": 0.311872661113739, "learning_rate": 0.0002, "epoch": 1.0521188504627375, "step": 1080}, {"loss": 1.7056, "grad_norm": 0.3276001513004303, "learning_rate": 0.0002, "epoch": 1.0618606916707258, "step": 1090}, {"loss": 1.7943, "grad_norm": 0.35227468609809875, "learning_rate": 0.0002, "epoch": 1.071602532878714, "step": 1100}, {"loss": 1.7079, "grad_norm": 0.3597564995288849, "learning_rate": 0.0002, "epoch": 1.0813443740867024, "step": 1110}, {"loss": 1.7635, "grad_norm": 0.3547225296497345, "learning_rate": 0.0002, "epoch": 1.0910862152946907, "step": 1120}, {"loss": 1.6859, "grad_norm": 0.3399786353111267, "learning_rate": 0.0002, "epoch": 1.100828056502679, "step": 1130}, {"loss": 1.7478, "grad_norm": 0.3309086263179779, "learning_rate": 0.0002, "epoch": 1.1105698977106673, "step": 1140}, {"loss": 1.7372, "grad_norm": 0.39330706000328064, "learning_rate": 0.0002, "epoch": 1.1203117389186557, "step": 1150}, {"loss": 1.6236, "grad_norm": 0.3628021776676178, "learning_rate": 0.0002, "epoch": 1.130053580126644, "step": 1160}, {"loss": 1.8022, "grad_norm": 0.32995012402534485, "learning_rate": 0.0002, "epoch": 1.1397954213346322, "step": 1170}, {"loss": 1.636, "grad_norm": 0.36292821168899536, "learning_rate": 0.0002, "epoch": 1.1495372625426206, "step": 1180}, {"loss": 1.7156, "grad_norm": 0.3470092713832855, "learning_rate": 0.0002, "epoch": 1.159279103750609, "step": 1190}, {"loss": 1.7001, "grad_norm": 0.3496156334877014, "learning_rate": 0.0002, "epoch": 1.169020944958597, "step": 1200}, {"loss": 1.716, "grad_norm": 0.3442084789276123, "learning_rate": 0.0002, "epoch": 1.1787627861665855, "step": 1210}, {"loss": 1.7763, "grad_norm": 0.34983909130096436, "learning_rate": 0.0002, "epoch": 1.1885046273745739, "step": 1220}, {"loss": 1.6964, "grad_norm": 0.36505937576293945, "learning_rate": 0.0002, "epoch": 1.198246468582562, "step": 1230}, {"loss": 1.7382, "grad_norm": 0.31624770164489746, "learning_rate": 0.0002, "epoch": 1.2079883097905504, "step": 1240}, {"loss": 1.7196, "grad_norm": 0.3528020679950714, "learning_rate": 0.0002, "epoch": 1.2177301509985388, "step": 1250}, {"loss": 1.752, "grad_norm": 0.29294025897979736, "learning_rate": 0.0002, "epoch": 1.2274719922065271, "step": 1260}, {"loss": 1.719, "grad_norm": 0.35048434138298035, "learning_rate": 0.0002, "epoch": 1.2372138334145153, "step": 1270}, {"loss": 1.6806, "grad_norm": 0.35224461555480957, "learning_rate": 0.0002, "epoch": 1.2469556746225037, "step": 1280}, {"loss": 1.6836, "grad_norm": 0.4041554629802704, "learning_rate": 0.0002, "epoch": 1.256697515830492, "step": 1290}, {"loss": 1.6999, "grad_norm": 0.3447791039943695, "learning_rate": 0.0002, "epoch": 1.2664393570384802, "step": 1300}, {"loss": 1.7601, "grad_norm": 0.3315333425998688, "learning_rate": 0.0002, "epoch": 1.2761811982464686, "step": 1310}, {"loss": 1.7325, "grad_norm": 0.3587741255760193, "learning_rate": 0.0002, "epoch": 1.285923039454457, "step": 1320}, {"loss": 1.6707, "grad_norm": 0.3704394996166229, "learning_rate": 0.0002, "epoch": 1.2956648806624451, "step": 1330}, {"loss": 1.7374, "grad_norm": 0.38131803274154663, "learning_rate": 0.0002, "epoch": 1.3054067218704335, "step": 1340}, {"loss": 1.5956, "grad_norm": 0.36109617352485657, "learning_rate": 0.0002, "epoch": 1.3151485630784219, "step": 1350}, {"loss": 1.682, "grad_norm": 0.37283554673194885, "learning_rate": 0.0002, "epoch": 1.32489040428641, "step": 1360}, {"loss": 1.7351, "grad_norm": 0.31808891892433167, "learning_rate": 0.0002, "epoch": 1.3346322454943984, "step": 1370}, {"loss": 1.8368, "grad_norm": 0.3370385766029358, "learning_rate": 0.0002, "epoch": 1.3443740867023868, "step": 1380}, {"loss": 1.6922, "grad_norm": 0.3568558394908905, "learning_rate": 0.0002, "epoch": 1.354115927910375, "step": 1390}, {"loss": 1.7301, "grad_norm": 0.3537410497665405, "learning_rate": 0.0002, "epoch": 1.3638577691183633, "step": 1400}, {"loss": 1.6534, "grad_norm": 0.3536544144153595, "learning_rate": 0.0002, "epoch": 1.3735996103263517, "step": 1410}, {"loss": 1.6829, "grad_norm": 0.3772895038127899, "learning_rate": 0.0002, "epoch": 1.38334145153434, "step": 1420}, {"loss": 1.7344, "grad_norm": 0.38079720735549927, "learning_rate": 0.0002, "epoch": 1.3930832927423282, "step": 1430}, {"loss": 1.7121, "grad_norm": 0.3811109662055969, "learning_rate": 0.0002, "epoch": 1.4028251339503166, "step": 1440}, {"loss": 1.6424, "grad_norm": 0.38586318492889404, "learning_rate": 0.0002, "epoch": 1.412566975158305, "step": 1450}, {"loss": 1.7438, "grad_norm": 0.3405744135379791, "learning_rate": 0.0002, "epoch": 1.4223088163662934, "step": 1460}, {"loss": 1.7483, "grad_norm": 0.39527642726898193, "learning_rate": 0.0002, "epoch": 1.4320506575742815, "step": 1470}, {"loss": 1.7927, "grad_norm": 0.4494728744029999, "learning_rate": 0.0002, "epoch": 1.44179249878227, "step": 1480}, {"loss": 1.5993, "grad_norm": 0.34068453311920166, "learning_rate": 0.0002, "epoch": 1.4515343399902583, "step": 1490}, {"loss": 1.73, "grad_norm": 0.36169710755348206, "learning_rate": 0.0002, "epoch": 1.4612761811982464, "step": 1500}, {"loss": 1.7189, "grad_norm": 0.31519418954849243, "learning_rate": 0.0002, "epoch": 1.4710180224062348, "step": 1510}, {"loss": 1.7533, "grad_norm": 0.35117292404174805, "learning_rate": 0.0002, "epoch": 1.4807598636142232, "step": 1520}, {"loss": 1.6662, "grad_norm": 0.40951141715049744, "learning_rate": 0.0002, "epoch": 1.4905017048222113, "step": 1530}, {"loss": 1.764, "grad_norm": 0.37542906403541565, "learning_rate": 0.0002, "epoch": 1.5002435460301997, "step": 1540}, {"loss": 1.6563, "grad_norm": 0.35395753383636475, "learning_rate": 0.0002, "epoch": 1.509985387238188, "step": 1550}, {"loss": 1.6517, "grad_norm": 0.35497018694877625, "learning_rate": 0.0002, "epoch": 1.5197272284461762, "step": 1560}, {"loss": 1.697, "grad_norm": 0.3693031072616577, "learning_rate": 0.0002, "epoch": 1.5294690696541646, "step": 1570}, {"loss": 1.7193, "grad_norm": 0.34013301134109497, "learning_rate": 0.0002, "epoch": 1.539210910862153, "step": 1580}, {"loss": 1.7364, "grad_norm": 0.37312784790992737, "learning_rate": 0.0002, "epoch": 1.5489527520701412, "step": 1590}, {"loss": 1.6698, "grad_norm": 0.357496440410614, "learning_rate": 0.0002, "epoch": 1.5586945932781295, "step": 1600}, {"loss": 1.7113, "grad_norm": 0.35192370414733887, "learning_rate": 0.0002, "epoch": 1.568436434486118, "step": 1610}, {"loss": 1.7505, "grad_norm": 0.34144821763038635, "learning_rate": 0.0002, "epoch": 1.578178275694106, "step": 1620}, {"loss": 1.6353, "grad_norm": 0.3320509195327759, "learning_rate": 0.0002, "epoch": 1.5879201169020944, "step": 1630}, {"loss": 1.6781, "grad_norm": 0.34178847074508667, "learning_rate": 0.0002, "epoch": 1.5976619581100828, "step": 1640}, {"loss": 1.7318, "grad_norm": 0.36567580699920654, "learning_rate": 0.0002, "epoch": 1.607403799318071, "step": 1650}, {"loss": 1.6971, "grad_norm": 0.35599812865257263, "learning_rate": 0.0002, "epoch": 1.6171456405260596, "step": 1660}, {"loss": 1.7256, "grad_norm": 0.33765384554862976, "learning_rate": 0.0002, "epoch": 1.6268874817340477, "step": 1670}, {"loss": 1.7396, "grad_norm": 0.33142679929733276, "learning_rate": 0.0002, "epoch": 1.636629322942036, "step": 1680}, {"loss": 1.7143, "grad_norm": 0.6959079504013062, "learning_rate": 0.0002, "epoch": 1.6463711641500245, "step": 1690}, {"loss": 1.7665, "grad_norm": 0.35073819756507874, "learning_rate": 0.0002, "epoch": 1.6561130053580126, "step": 1700}, {"loss": 1.7571, "grad_norm": 0.3461478352546692, "learning_rate": 0.0002, "epoch": 1.665854846566001, "step": 1710}, {"loss": 1.6608, "grad_norm": 0.3697752058506012, "learning_rate": 0.0002, "epoch": 1.6755966877739894, "step": 1720}, {"loss": 1.729, "grad_norm": 0.3755154609680176, "learning_rate": 0.0002, "epoch": 1.6853385289819776, "step": 1730}, {"loss": 1.6618, "grad_norm": 0.33977627754211426, "learning_rate": 0.0002, "epoch": 1.695080370189966, "step": 1740}, {"loss": 1.7207, "grad_norm": 0.4001041650772095, "learning_rate": 0.0002, "epoch": 1.7048222113979543, "step": 1750}, {"loss": 1.6756, "grad_norm": 0.36998286843299866, "learning_rate": 0.0002, "epoch": 1.7145640526059425, "step": 1760}, {"loss": 1.7506, "grad_norm": 0.39944565296173096, "learning_rate": 0.0002, "epoch": 1.7243058938139308, "step": 1770}, {"loss": 1.6725, "grad_norm": 0.4002859890460968, "learning_rate": 0.0002, "epoch": 1.7340477350219192, "step": 1780}, {"loss": 1.768, "grad_norm": 0.33336859941482544, "learning_rate": 0.0002, "epoch": 1.7437895762299074, "step": 1790}, {"loss": 1.7816, "grad_norm": 0.35853952169418335, "learning_rate": 0.0002, "epoch": 1.7535314174378958, "step": 1800}, {"loss": 1.7135, "grad_norm": 0.35876700282096863, "learning_rate": 0.0002, "epoch": 1.7632732586458841, "step": 1810}, {"loss": 1.7327, "grad_norm": 0.3497968912124634, "learning_rate": 0.0002, "epoch": 1.7730150998538723, "step": 1820}, {"loss": 1.7128, "grad_norm": 0.33182016015052795, "learning_rate": 0.0002, "epoch": 1.7827569410618607, "step": 1830}, {"loss": 1.7594, "grad_norm": 0.33359771966934204, "learning_rate": 0.0002, "epoch": 1.792498782269849, "step": 1840}, {"loss": 1.8611, "grad_norm": 0.38070961833000183, "learning_rate": 0.0002, "epoch": 1.8022406234778372, "step": 1850}, {"loss": 1.7576, "grad_norm": 0.34111160039901733, "learning_rate": 0.0002, "epoch": 1.8119824646858256, "step": 1860}, {"loss": 1.7533, "grad_norm": 0.4439302980899811, "learning_rate": 0.0002, "epoch": 1.821724305893814, "step": 1870}, {"loss": 1.7931, "grad_norm": 0.37065210938453674, "learning_rate": 0.0002, "epoch": 1.8314661471018021, "step": 1880}, {"loss": 1.7392, "grad_norm": 0.33630406856536865, "learning_rate": 0.0002, "epoch": 1.8412079883097907, "step": 1890}, {"loss": 1.6369, "grad_norm": 0.334553986787796, "learning_rate": 0.0002, "epoch": 1.8509498295177789, "step": 1900}, {"loss": 1.7271, "grad_norm": 0.3603808879852295, "learning_rate": 0.0002, "epoch": 1.860691670725767, "step": 1910}, {"loss": 1.6777, "grad_norm": 0.4307343363761902, "learning_rate": 0.0002, "epoch": 1.8704335119337556, "step": 1920}, {"loss": 1.75, "grad_norm": 0.455602765083313, "learning_rate": 0.0002, "epoch": 1.8801753531417438, "step": 1930}, {"loss": 1.6898, "grad_norm": 0.35242316126823425, "learning_rate": 0.0002, "epoch": 1.8899171943497322, "step": 1940}, {"loss": 1.7152, "grad_norm": 0.3589116632938385, "learning_rate": 0.0002, "epoch": 1.8996590355577205, "step": 1950}, {"loss": 1.7125, "grad_norm": 0.3540741801261902, "learning_rate": 0.0002, "epoch": 1.9094008767657087, "step": 1960}, {"loss": 1.6873, "grad_norm": 0.3547612428665161, "learning_rate": 0.0002, "epoch": 1.919142717973697, "step": 1970}, {"loss": 1.6995, "grad_norm": 0.3485773503780365, "learning_rate": 0.0002, "epoch": 1.9288845591816854, "step": 1980}, {"loss": 1.7301, "grad_norm": 0.3560304641723633, "learning_rate": 0.0002, "epoch": 1.9386264003896736, "step": 1990}, {"loss": 1.748, "grad_norm": 0.33299335837364197, "learning_rate": 0.0002, "epoch": 1.948368241597662, "step": 2000}, {"loss": 1.7397, "grad_norm": 0.35622233152389526, "learning_rate": 0.0002, "epoch": 1.9581100828056504, "step": 2010}, {"loss": 1.7201, "grad_norm": 0.3681301474571228, "learning_rate": 0.0002, "epoch": 1.9678519240136385, "step": 2020}, {"loss": 1.717, "grad_norm": 0.36158084869384766, "learning_rate": 0.0002, "epoch": 1.9775937652216269, "step": 2030}, {"loss": 1.6332, "grad_norm": 0.32560569047927856, "learning_rate": 0.0002, "epoch": 1.9873356064296153, "step": 2040}, {"loss": 1.6958, "grad_norm": 0.37404149770736694, "learning_rate": 0.0002, "epoch": 1.9970774476376034, "step": 2050}, {"eval_loss": 1.8119343519210815, "eval_runtime": 96.0045, "eval_samples_per_second": 5.281, "eval_steps_per_second": 0.667, "epoch": 2.0, "step": 2053}, {"loss": 1.657, "grad_norm": 0.374188631772995, "learning_rate": 0.0002, "epoch": 2.006819288845592, "step": 2060}, {"loss": 1.5655, "grad_norm": 0.421764075756073, "learning_rate": 0.0002, "epoch": 2.01656113005358, "step": 2070}, {"loss": 1.548, "grad_norm": 0.43841829895973206, "learning_rate": 0.0002, "epoch": 2.0263029712615683, "step": 2080}, {"loss": 1.6326, "grad_norm": 0.42298218607902527, "learning_rate": 0.0002, "epoch": 2.036044812469557, "step": 2090}, {"loss": 1.5883, "grad_norm": 0.43669602274894714, "learning_rate": 0.0002, "epoch": 2.045786653677545, "step": 2100}, {"loss": 1.6143, "grad_norm": 0.4080469012260437, "learning_rate": 0.0002, "epoch": 2.0555284948855332, "step": 2110}, {"loss": 1.5591, "grad_norm": 0.483192503452301, "learning_rate": 0.0002, "epoch": 2.065270336093522, "step": 2120}, {"loss": 1.6492, "grad_norm": 0.44427400827407837, "learning_rate": 0.0002, "epoch": 2.07501217730151, "step": 2130}, {"loss": 1.5845, "grad_norm": 0.48835131525993347, "learning_rate": 0.0002, "epoch": 2.084754018509498, "step": 2140}, {"loss": 1.5617, "grad_norm": 0.42733684182167053, "learning_rate": 0.0002, "epoch": 2.0944958597174868, "step": 2150}, {"loss": 1.5562, "grad_norm": 0.4258694648742676, "learning_rate": 0.0002, "epoch": 2.104237700925475, "step": 2160}, {"loss": 1.517, "grad_norm": 0.5164985656738281, "learning_rate": 0.0002, "epoch": 2.113979542133463, "step": 2170}, {"loss": 1.6393, "grad_norm": 0.4279228150844574, "learning_rate": 0.0002, "epoch": 2.1237213833414517, "step": 2180}, {"loss": 1.5895, "grad_norm": 0.48209506273269653, "learning_rate": 0.0002, "epoch": 2.13346322454944, "step": 2190}, {"loss": 1.6142, "grad_norm": 0.4071785509586334, "learning_rate": 0.0002, "epoch": 2.143205065757428, "step": 2200}, {"loss": 1.5469, "grad_norm": 0.4629398584365845, "learning_rate": 0.0002, "epoch": 2.1529469069654166, "step": 2210}, {"loss": 1.56, "grad_norm": 0.44390997290611267, "learning_rate": 0.0002, "epoch": 2.1626887481734047, "step": 2220}, {"loss": 1.5395, "grad_norm": 0.46886971592903137, "learning_rate": 0.0002, "epoch": 2.172430589381393, "step": 2230}, {"loss": 1.6108, "grad_norm": 0.43745434284210205, "learning_rate": 0.0002, "epoch": 2.1821724305893815, "step": 2240}, {"loss": 1.5416, "grad_norm": 0.42737245559692383, "learning_rate": 0.0002, "epoch": 2.1919142717973696, "step": 2250}, {"loss": 1.627, "grad_norm": 0.5028428435325623, "learning_rate": 0.0002, "epoch": 2.201656113005358, "step": 2260}, {"loss": 1.6148, "grad_norm": 0.48987212777137756, "learning_rate": 0.0002, "epoch": 2.2113979542133464, "step": 2270}, {"loss": 1.6107, "grad_norm": 0.48186370730400085, "learning_rate": 0.0002, "epoch": 2.2211397954213346, "step": 2280}, {"loss": 1.6657, "grad_norm": 0.4417429566383362, "learning_rate": 0.0002, "epoch": 2.2308816366293227, "step": 2290}, {"loss": 1.595, "grad_norm": 0.4757710099220276, "learning_rate": 0.0002, "epoch": 2.2406234778373113, "step": 2300}, {"loss": 1.591, "grad_norm": 0.44449448585510254, "learning_rate": 0.0002, "epoch": 2.2503653190452995, "step": 2310}, {"loss": 1.5742, "grad_norm": 0.5070863962173462, "learning_rate": 0.0002, "epoch": 2.260107160253288, "step": 2320}, {"loss": 1.5831, "grad_norm": 0.4967133700847626, "learning_rate": 0.0002, "epoch": 2.269849001461276, "step": 2330}, {"loss": 1.5857, "grad_norm": 0.5110220909118652, "learning_rate": 0.0002, "epoch": 2.2795908426692644, "step": 2340}, {"loss": 1.6266, "grad_norm": 0.47984135150909424, "learning_rate": 0.0002, "epoch": 2.289332683877253, "step": 2350}, {"loss": 1.5927, "grad_norm": 0.5005794763565063, "learning_rate": 0.0002, "epoch": 2.299074525085241, "step": 2360}, {"loss": 1.6131, "grad_norm": 0.4991425573825836, "learning_rate": 0.0002, "epoch": 2.3088163662932293, "step": 2370}, {"loss": 1.5386, "grad_norm": 0.4948616623878479, "learning_rate": 0.0002, "epoch": 2.318558207501218, "step": 2380}, {"loss": 1.5769, "grad_norm": 0.4533160328865051, "learning_rate": 0.0002, "epoch": 2.328300048709206, "step": 2390}, {"loss": 1.5856, "grad_norm": 0.5871071219444275, "learning_rate": 0.0002, "epoch": 2.338041889917194, "step": 2400}, {"loss": 1.591, "grad_norm": 0.5048075914382935, "learning_rate": 0.0002, "epoch": 2.347783731125183, "step": 2410}, {"loss": 1.6165, "grad_norm": 0.4973750412464142, "learning_rate": 0.0002, "epoch": 2.357525572333171, "step": 2420}, {"loss": 1.589, "grad_norm": 0.48294538259506226, "learning_rate": 0.0002, "epoch": 2.367267413541159, "step": 2430}, {"loss": 1.5782, "grad_norm": 0.7180454134941101, "learning_rate": 0.0002, "epoch": 2.3770092547491477, "step": 2440}, {"loss": 1.56, "grad_norm": 0.4627632796764374, "learning_rate": 0.0002, "epoch": 2.386751095957136, "step": 2450}, {"loss": 1.586, "grad_norm": 0.4834378957748413, "learning_rate": 0.0002, "epoch": 2.396492937165124, "step": 2460}, {"loss": 1.6145, "grad_norm": 0.5173670649528503, "learning_rate": 0.0002, "epoch": 2.4062347783731126, "step": 2470}, {"loss": 1.5464, "grad_norm": 0.49652737379074097, "learning_rate": 0.0002, "epoch": 2.4159766195811008, "step": 2480}, {"loss": 1.6977, "grad_norm": 0.47052669525146484, "learning_rate": 0.0002, "epoch": 2.4257184607890894, "step": 2490}, {"loss": 1.6215, "grad_norm": 0.5188006162643433, "learning_rate": 0.0002, "epoch": 2.4354603019970775, "step": 2500}, {"loss": 1.5512, "grad_norm": 0.5010119676589966, "learning_rate": 0.0002, "epoch": 2.4452021432050657, "step": 2510}, {"loss": 1.6403, "grad_norm": 0.4765235483646393, "learning_rate": 0.0002, "epoch": 2.4549439844130543, "step": 2520}, {"loss": 1.5907, "grad_norm": 0.5292699337005615, "learning_rate": 0.0002, "epoch": 2.4646858256210424, "step": 2530}, {"loss": 1.5866, "grad_norm": 0.48555099964141846, "learning_rate": 0.0002, "epoch": 2.4744276668290306, "step": 2540}, {"loss": 1.5361, "grad_norm": 0.4764043092727661, "learning_rate": 0.0002, "epoch": 2.484169508037019, "step": 2550}, {"loss": 1.6545, "grad_norm": 0.47839659452438354, "learning_rate": 0.0002, "epoch": 2.4939113492450073, "step": 2560}, {"loss": 1.6269, "grad_norm": 0.4514436721801758, "learning_rate": 0.0002, "epoch": 2.5036531904529955, "step": 2570}, {"loss": 1.6389, "grad_norm": 0.5681955218315125, "learning_rate": 0.0002, "epoch": 2.513395031660984, "step": 2580}, {"loss": 1.5976, "grad_norm": 0.49655985832214355, "learning_rate": 0.0002, "epoch": 2.5231368728689723, "step": 2590}, {"loss": 1.6052, "grad_norm": 0.5077657103538513, "learning_rate": 0.0002, "epoch": 2.5328787140769604, "step": 2600}, {"loss": 1.5658, "grad_norm": 0.5643279552459717, "learning_rate": 0.0002, "epoch": 2.542620555284949, "step": 2610}, {"loss": 1.5455, "grad_norm": 0.4715031087398529, "learning_rate": 0.0002, "epoch": 2.552362396492937, "step": 2620}, {"loss": 1.5907, "grad_norm": 0.528400719165802, "learning_rate": 0.0002, "epoch": 2.5621042377009253, "step": 2630}, {"loss": 1.5452, "grad_norm": 0.49469611048698425, "learning_rate": 0.0002, "epoch": 2.571846078908914, "step": 2640}, {"loss": 1.5904, "grad_norm": 0.4567806124687195, "learning_rate": 0.0002, "epoch": 2.581587920116902, "step": 2650}, {"loss": 1.6242, "grad_norm": 0.5357107520103455, "learning_rate": 0.0002, "epoch": 2.5913297613248902, "step": 2660}, {"loss": 1.6078, "grad_norm": 0.46977677941322327, "learning_rate": 0.0002, "epoch": 2.601071602532879, "step": 2670}, {"loss": 1.6994, "grad_norm": 0.6626771092414856, "learning_rate": 0.0002, "epoch": 2.610813443740867, "step": 2680}, {"loss": 1.5888, "grad_norm": 0.4587472081184387, "learning_rate": 0.0002, "epoch": 2.620555284948855, "step": 2690}, {"loss": 1.6002, "grad_norm": 0.4816797077655792, "learning_rate": 0.0002, "epoch": 2.6302971261568437, "step": 2700}, {"loss": 1.5701, "grad_norm": 0.4856809675693512, "learning_rate": 0.0002, "epoch": 2.640038967364832, "step": 2710}, {"loss": 1.563, "grad_norm": 0.46010780334472656, "learning_rate": 0.0002, "epoch": 2.64978080857282, "step": 2720}, {"loss": 1.5374, "grad_norm": 0.4637954533100128, "learning_rate": 0.0002, "epoch": 2.6595226497808087, "step": 2730}, {"loss": 1.6493, "grad_norm": 0.5954997539520264, "learning_rate": 0.0002, "epoch": 2.669264490988797, "step": 2740}, {"loss": 1.5795, "grad_norm": 0.5071861743927002, "learning_rate": 0.0002, "epoch": 2.679006332196785, "step": 2750}, {"loss": 1.573, "grad_norm": 0.5415477156639099, "learning_rate": 0.0002, "epoch": 2.6887481734047736, "step": 2760}, {"loss": 1.5476, "grad_norm": 0.5618549585342407, "learning_rate": 0.0002, "epoch": 2.6984900146127617, "step": 2770}, {"loss": 1.608, "grad_norm": 0.49338817596435547, "learning_rate": 0.0002, "epoch": 2.70823185582075, "step": 2780}, {"loss": 1.6529, "grad_norm": 0.5149586796760559, "learning_rate": 0.0002, "epoch": 2.7179736970287385, "step": 2790}, {"loss": 1.6279, "grad_norm": 0.6247242093086243, "learning_rate": 0.0002, "epoch": 2.7277155382367266, "step": 2800}, {"loss": 1.4655, "grad_norm": 0.4749542474746704, "learning_rate": 0.0002, "epoch": 2.737457379444715, "step": 2810}, {"loss": 1.5984, "grad_norm": 0.4979191720485687, "learning_rate": 0.0002, "epoch": 2.7471992206527034, "step": 2820}, {"loss": 1.6377, "grad_norm": 0.4885074198246002, "learning_rate": 0.0002, "epoch": 2.7569410618606915, "step": 2830}, {"loss": 1.6529, "grad_norm": 0.5047747492790222, "learning_rate": 0.0002, "epoch": 2.76668290306868, "step": 2840}, {"loss": 1.6574, "grad_norm": 0.5280140042304993, "learning_rate": 0.0002, "epoch": 2.7764247442766683, "step": 2850}, {"loss": 1.5639, "grad_norm": 0.477668434381485, "learning_rate": 0.0002, "epoch": 2.7861665854846565, "step": 2860}, {"loss": 1.5923, "grad_norm": 0.4816327393054962, "learning_rate": 0.0002, "epoch": 2.795908426692645, "step": 2870}, {"loss": 1.6377, "grad_norm": 0.523259162902832, "learning_rate": 0.0002, "epoch": 2.805650267900633, "step": 2880}, {"loss": 1.5779, "grad_norm": 0.5045270919799805, "learning_rate": 0.0002, "epoch": 2.8153921091086214, "step": 2890}, {"loss": 1.5915, "grad_norm": 0.47986042499542236, "learning_rate": 0.0002, "epoch": 2.82513395031661, "step": 2900}, {"loss": 1.5997, "grad_norm": 0.4858797490596771, "learning_rate": 0.0002, "epoch": 2.834875791524598, "step": 2910}, {"loss": 1.5664, "grad_norm": 0.5261512398719788, "learning_rate": 0.0002, "epoch": 2.8446176327325867, "step": 2920}, {"loss": 1.5775, "grad_norm": 0.630550742149353, "learning_rate": 0.0002, "epoch": 2.854359473940575, "step": 2930}, {"loss": 1.5889, "grad_norm": 0.49119752645492554, "learning_rate": 0.0002, "epoch": 2.864101315148563, "step": 2940}, {"loss": 1.6033, "grad_norm": 0.4779070317745209, "learning_rate": 0.0002, "epoch": 2.8738431563565516, "step": 2950}, {"loss": 1.6353, "grad_norm": 0.5059782266616821, "learning_rate": 0.0002, "epoch": 2.88358499756454, "step": 2960}, {"loss": 1.6403, "grad_norm": 0.5466655492782593, "learning_rate": 0.0002, "epoch": 2.893326838772528, "step": 2970}, {"loss": 1.619, "grad_norm": 0.4865640103816986, "learning_rate": 0.0002, "epoch": 2.9030686799805165, "step": 2980}, {"loss": 1.5712, "grad_norm": 0.6101235151290894, "learning_rate": 0.0002, "epoch": 2.9128105211885047, "step": 2990}, {"loss": 1.6216, "grad_norm": 0.9112305641174316, "learning_rate": 0.0002, "epoch": 2.922552362396493, "step": 3000}, {"loss": 1.6493, "grad_norm": 0.3938814103603363, "learning_rate": 0.0002, "epoch": 2.9322942036044815, "step": 3010}, {"loss": 1.5446, "grad_norm": 0.5500800609588623, "learning_rate": 0.0002, "epoch": 2.9420360448124696, "step": 3020}, {"loss": 1.5626, "grad_norm": 0.5346390604972839, "learning_rate": 0.0002, "epoch": 2.9517778860204578, "step": 3030}, {"loss": 1.5783, "grad_norm": 0.5245014429092407, "learning_rate": 0.0002, "epoch": 2.9615197272284464, "step": 3040}, {"loss": 1.5549, "grad_norm": 0.4906884431838989, "learning_rate": 0.0002, "epoch": 2.9712615684364345, "step": 3050}, {"loss": 1.5796, "grad_norm": 0.47086769342422485, "learning_rate": 0.0002, "epoch": 2.9810034096444227, "step": 3060}, {"loss": 1.5917, "grad_norm": 0.5290229320526123, "learning_rate": 0.0002, "epoch": 2.9907452508524113, "step": 3070}, {"eval_loss": 1.8463934659957886, "eval_runtime": 56.2401, "eval_samples_per_second": 9.015, "eval_steps_per_second": 1.138, "epoch": 2.9995129079396006, "step": 3079}, {"loss": 1.5744, "grad_norm": 0.49992576241493225, "learning_rate": 0.0002, "epoch": 3.0004870920603994, "step": 3080}, {"loss": 1.4125, "grad_norm": 0.8242783546447754, "learning_rate": 0.0002, "epoch": 3.0102289332683876, "step": 3090}, {"loss": 1.394, "grad_norm": 0.6330569386482239, "learning_rate": 0.0002, "epoch": 3.019970774476376, "step": 3100}, {"loss": 1.4942, "grad_norm": 0.566097617149353, "learning_rate": 0.0002, "epoch": 3.0297126156843643, "step": 3110}, {"loss": 1.4365, "grad_norm": 0.6337586045265198, "learning_rate": 0.0002, "epoch": 3.0394544568923525, "step": 3120}, {"loss": 1.3916, "grad_norm": 0.7339403033256531, "learning_rate": 0.0002, "epoch": 3.049196298100341, "step": 3130}, {"loss": 1.4617, "grad_norm": 0.7187346816062927, "learning_rate": 0.0002, "epoch": 3.0589381393083293, "step": 3140}, {"loss": 1.3453, "grad_norm": 0.7116255760192871, "learning_rate": 0.0002, "epoch": 3.0686799805163174, "step": 3150}, {"loss": 1.4452, "grad_norm": 0.6493807435035706, "learning_rate": 0.0002, "epoch": 3.078421821724306, "step": 3160}, {"loss": 1.351, "grad_norm": 0.6777266263961792, "learning_rate": 0.0002, "epoch": 3.088163662932294, "step": 3170}, {"loss": 1.4362, "grad_norm": 0.6342006325721741, "learning_rate": 0.0002, "epoch": 3.0979055041402823, "step": 3180}, {"loss": 1.4748, "grad_norm": 0.6608964204788208, "learning_rate": 0.0002, "epoch": 3.107647345348271, "step": 3190}, {"loss": 1.375, "grad_norm": 0.7230247259140015, "learning_rate": 0.0002, "epoch": 3.117389186556259, "step": 3200}, {"loss": 1.4049, "grad_norm": 0.650368332862854, "learning_rate": 0.0002, "epoch": 3.1271310277642472, "step": 3210}, {"loss": 1.409, "grad_norm": 0.7319342494010925, "learning_rate": 0.0002, "epoch": 3.136872868972236, "step": 3220}, {"loss": 1.3872, "grad_norm": 0.7159963846206665, "learning_rate": 0.0002, "epoch": 3.146614710180224, "step": 3230}, {"loss": 1.5076, "grad_norm": 0.8905230164527893, "learning_rate": 0.0002, "epoch": 3.156356551388212, "step": 3240}, {"loss": 1.3161, "grad_norm": 0.6920804381370544, "learning_rate": 0.0002, "epoch": 3.1660983925962007, "step": 3250}, {"loss": 1.3786, "grad_norm": 0.6782063841819763, "learning_rate": 0.0002, "epoch": 3.175840233804189, "step": 3260}, {"loss": 1.5153, "grad_norm": 0.735325276851654, "learning_rate": 0.0002, "epoch": 3.1855820750121775, "step": 3270}, {"loss": 1.4027, "grad_norm": 0.6657978296279907, "learning_rate": 0.0002, "epoch": 3.1953239162201656, "step": 3280}, {"loss": 1.3456, "grad_norm": 0.771315336227417, "learning_rate": 0.0002, "epoch": 3.205065757428154, "step": 3290}, {"loss": 1.3236, "grad_norm": 0.6492983102798462, "learning_rate": 0.0002, "epoch": 3.2148075986361424, "step": 3300}, {"loss": 1.4125, "grad_norm": 0.7513770461082458, "learning_rate": 0.0002, "epoch": 3.2245494398441306, "step": 3310}, {"loss": 1.4032, "grad_norm": 0.7091423869132996, "learning_rate": 0.0002, "epoch": 3.2342912810521187, "step": 3320}, {"loss": 1.4585, "grad_norm": 0.6663975119590759, "learning_rate": 0.0002, "epoch": 3.2440331222601073, "step": 3330}, {"loss": 1.3968, "grad_norm": 0.6813122034072876, "learning_rate": 0.0002, "epoch": 3.2537749634680955, "step": 3340}, {"loss": 1.3681, "grad_norm": 0.6602569818496704, "learning_rate": 0.0002, "epoch": 3.2635168046760836, "step": 3350}, {"loss": 1.4533, "grad_norm": 0.718270480632782, "learning_rate": 0.0002, "epoch": 3.2732586458840722, "step": 3360}, {"loss": 1.4076, "grad_norm": 0.6884173154830933, "learning_rate": 0.0002, "epoch": 3.2830004870920604, "step": 3370}, {"loss": 1.4144, "grad_norm": 0.7039775848388672, "learning_rate": 0.0002, "epoch": 3.2927423283000485, "step": 3380}, {"loss": 1.5077, "grad_norm": 0.7444299459457397, "learning_rate": 0.0002, "epoch": 3.302484169508037, "step": 3390}, {"loss": 1.4255, "grad_norm": 0.7187064290046692, "learning_rate": 0.0002, "epoch": 3.3122260107160253, "step": 3400}, {"loss": 1.3684, "grad_norm": 0.599396288394928, "learning_rate": 0.0002, "epoch": 3.3219678519240134, "step": 3410}, {"loss": 1.4819, "grad_norm": 0.7670390009880066, "learning_rate": 0.0002, "epoch": 3.331709693132002, "step": 3420}, {"loss": 1.4411, "grad_norm": 0.6654478311538696, "learning_rate": 0.0002, "epoch": 3.34145153433999, "step": 3430}, {"loss": 1.4257, "grad_norm": 0.6644385457038879, "learning_rate": 0.0002, "epoch": 3.351193375547979, "step": 3440}, {"loss": 1.4508, "grad_norm": 0.6974098086357117, "learning_rate": 0.0002, "epoch": 3.360935216755967, "step": 3450}, {"loss": 1.3807, "grad_norm": 0.7350399494171143, "learning_rate": 0.0002, "epoch": 3.370677057963955, "step": 3460}, {"loss": 1.4176, "grad_norm": 0.714721143245697, "learning_rate": 0.0002, "epoch": 3.3804188991719437, "step": 3470}, {"loss": 1.4325, "grad_norm": 0.7006027698516846, "learning_rate": 0.0002, "epoch": 3.390160740379932, "step": 3480}, {"loss": 1.4888, "grad_norm": 0.6767925024032593, "learning_rate": 0.0002, "epoch": 3.39990258158792, "step": 3490}, {"loss": 1.4116, "grad_norm": 0.6721355319023132, "learning_rate": 0.0002, "epoch": 3.4096444227959086, "step": 3500}, {"loss": 1.443, "grad_norm": 0.6845725178718567, "learning_rate": 0.0002, "epoch": 3.419386264003897, "step": 3510}, {"loss": 1.4832, "grad_norm": 0.6882196664810181, "learning_rate": 0.0002, "epoch": 3.429128105211885, "step": 3520}, {"loss": 1.4962, "grad_norm": 0.7663240432739258, "learning_rate": 0.0002, "epoch": 3.4388699464198735, "step": 3530}, {"loss": 1.4644, "grad_norm": 0.6304219365119934, "learning_rate": 0.0002, "epoch": 3.4486117876278617, "step": 3540}, {"loss": 1.4918, "grad_norm": 0.668678879737854, "learning_rate": 0.0002, "epoch": 3.45835362883585, "step": 3550}, {"loss": 1.4874, "grad_norm": 0.7526912093162537, "learning_rate": 0.0002, "epoch": 3.4680954700438384, "step": 3560}, {"loss": 1.4249, "grad_norm": 1.089495301246643, "learning_rate": 0.0002, "epoch": 3.4778373112518266, "step": 3570}, {"loss": 1.3871, "grad_norm": 0.7282902002334595, "learning_rate": 0.0002, "epoch": 3.4875791524598148, "step": 3580}, {"loss": 1.5077, "grad_norm": 0.6540156602859497, "learning_rate": 0.0002, "epoch": 3.4973209936678034, "step": 3590}, {"loss": 1.4367, "grad_norm": 0.6449568867683411, "learning_rate": 0.0002, "epoch": 3.5070628348757915, "step": 3600}, {"loss": 1.4532, "grad_norm": 0.7262216210365295, "learning_rate": 0.0002, "epoch": 3.5168046760837797, "step": 3610}, {"loss": 1.4374, "grad_norm": 0.6048615574836731, "learning_rate": 0.0002, "epoch": 3.5265465172917683, "step": 3620}, {"loss": 1.3877, "grad_norm": 0.6780537366867065, "learning_rate": 0.0002, "epoch": 3.5362883584997564, "step": 3630}, {"loss": 1.422, "grad_norm": 0.6851925253868103, "learning_rate": 0.0002, "epoch": 3.5460301997077446, "step": 3640}, {"loss": 1.3425, "grad_norm": 0.6530634164810181, "learning_rate": 0.0002, "epoch": 3.555772040915733, "step": 3650}, {"loss": 1.4879, "grad_norm": 0.7193992733955383, "learning_rate": 0.0002, "epoch": 3.5655138821237213, "step": 3660}, {"loss": 1.4555, "grad_norm": 0.767496645450592, "learning_rate": 0.0002, "epoch": 3.5752557233317095, "step": 3670}, {"loss": 1.4824, "grad_norm": 0.6912919282913208, "learning_rate": 0.0002, "epoch": 3.584997564539698, "step": 3680}, {"loss": 1.4497, "grad_norm": 0.7383436560630798, "learning_rate": 0.0002, "epoch": 3.5947394057476862, "step": 3690}, {"loss": 1.4822, "grad_norm": 0.6746662855148315, "learning_rate": 0.0002, "epoch": 3.6044812469556744, "step": 3700}, {"loss": 1.4904, "grad_norm": 0.6885138750076294, "learning_rate": 0.0002, "epoch": 3.614223088163663, "step": 3710}, {"loss": 1.4044, "grad_norm": 0.6694392561912537, "learning_rate": 0.0002, "epoch": 3.623964929371651, "step": 3720}, {"loss": 1.3719, "grad_norm": 0.812358021736145, "learning_rate": 0.0002, "epoch": 3.6337067705796393, "step": 3730}, {"loss": 1.4603, "grad_norm": 0.7267130017280579, "learning_rate": 0.0002, "epoch": 3.643448611787628, "step": 3740}, {"loss": 1.4574, "grad_norm": 0.6958749294281006, "learning_rate": 0.0002, "epoch": 3.653190452995616, "step": 3750}, {"loss": 1.4346, "grad_norm": 0.6805673241615295, "learning_rate": 0.0002, "epoch": 3.6629322942036042, "step": 3760}, {"loss": 1.4338, "grad_norm": 0.7184410095214844, "learning_rate": 0.0002, "epoch": 3.672674135411593, "step": 3770}, {"loss": 1.3935, "grad_norm": 0.7716330289840698, "learning_rate": 0.0002, "epoch": 3.682415976619581, "step": 3780}, {"loss": 1.384, "grad_norm": 0.6675831079483032, "learning_rate": 0.0002, "epoch": 3.6921578178275696, "step": 3790}, {"loss": 1.401, "grad_norm": 0.6480095386505127, "learning_rate": 0.0002, "epoch": 3.7018996590355577, "step": 3800}, {"loss": 1.5303, "grad_norm": 0.6559418439865112, "learning_rate": 0.0002, "epoch": 3.711641500243546, "step": 3810}, {"loss": 1.4341, "grad_norm": 0.6596545577049255, "learning_rate": 0.0002, "epoch": 3.7213833414515345, "step": 3820}, {"loss": 1.4508, "grad_norm": 0.7172950506210327, "learning_rate": 0.0002, "epoch": 3.7311251826595226, "step": 3830}, {"loss": 1.446, "grad_norm": 0.796148419380188, "learning_rate": 0.0002, "epoch": 3.740867023867511, "step": 3840}, {"loss": 1.4992, "grad_norm": 0.6600322723388672, "learning_rate": 0.0002, "epoch": 3.7506088650754994, "step": 3850}, {"loss": 1.4201, "grad_norm": 0.6776387691497803, "learning_rate": 0.0002, "epoch": 3.7603507062834876, "step": 3860}, {"loss": 1.3893, "grad_norm": 0.7768304347991943, "learning_rate": 0.0002, "epoch": 3.770092547491476, "step": 3870}, {"loss": 1.4886, "grad_norm": 1.0579794645309448, "learning_rate": 0.0002, "epoch": 3.7798343886994643, "step": 3880}, {"loss": 1.4556, "grad_norm": 0.6757252812385559, "learning_rate": 0.0002, "epoch": 3.7895762299074525, "step": 3890}, {"loss": 1.4647, "grad_norm": 0.6706996560096741, "learning_rate": 0.0002, "epoch": 3.799318071115441, "step": 3900}, {"loss": 1.4104, "grad_norm": 0.7026948928833008, "learning_rate": 0.0002, "epoch": 3.809059912323429, "step": 3910}, {"loss": 1.5487, "grad_norm": 0.6437768340110779, "learning_rate": 0.0002, "epoch": 3.8188017535314174, "step": 3920}, {"loss": 1.4678, "grad_norm": 0.7015706300735474, "learning_rate": 0.0002, "epoch": 3.828543594739406, "step": 3930}, {"loss": 1.4891, "grad_norm": 0.7049482464790344, "learning_rate": 0.0002, "epoch": 3.838285435947394, "step": 3940}, {"loss": 1.4208, "grad_norm": 0.6533724665641785, "learning_rate": 0.0002, "epoch": 3.8480272771553823, "step": 3950}, {"loss": 1.4435, "grad_norm": 0.7312499284744263, "learning_rate": 0.0002, "epoch": 3.857769118363371, "step": 3960}, {"loss": 1.3886, "grad_norm": 0.6858801245689392, "learning_rate": 0.0002, "epoch": 3.867510959571359, "step": 3970}, {"loss": 1.4423, "grad_norm": 0.770423173904419, "learning_rate": 0.0002, "epoch": 3.877252800779347, "step": 3980}, {"loss": 1.5029, "grad_norm": 0.6987539529800415, "learning_rate": 0.0002, "epoch": 3.886994641987336, "step": 3990}, {"loss": 1.4791, "grad_norm": 0.7072722315788269, "learning_rate": 0.0002, "epoch": 3.896736483195324, "step": 4000}, {"loss": 1.528, "grad_norm": 0.6492931842803955, "learning_rate": 0.0002, "epoch": 3.906478324403312, "step": 4010}, {"loss": 1.3824, "grad_norm": 0.7716232538223267, "learning_rate": 0.0002, "epoch": 3.9162201656113007, "step": 4020}, {"loss": 1.4758, "grad_norm": 0.722949743270874, "learning_rate": 0.0002, "epoch": 3.925962006819289, "step": 4030}, {"loss": 1.3914, "grad_norm": 0.7434365749359131, "learning_rate": 0.0002, "epoch": 3.935703848027277, "step": 4040}, {"loss": 1.4763, "grad_norm": 0.6691509485244751, "learning_rate": 0.0002, "epoch": 3.9454456892352656, "step": 4050}, {"loss": 1.4555, "grad_norm": 0.6850284337997437, "learning_rate": 0.0002, "epoch": 3.9551875304432538, "step": 4060}, {"loss": 1.5275, "grad_norm": 0.6954452991485596, "learning_rate": 0.0002, "epoch": 3.964929371651242, "step": 4070}, {"loss": 1.417, "grad_norm": 0.9316364526748657, "learning_rate": 0.0002, "epoch": 3.9746712128592305, "step": 4080}, {"loss": 1.4532, "grad_norm": 0.6908289194107056, "learning_rate": 0.0002, "epoch": 3.9844130540672187, "step": 4090}, {"loss": 1.4404, "grad_norm": 0.666782021522522, "learning_rate": 0.0002, "epoch": 3.994154895275207, "step": 4100}, {"eval_loss": 1.9233275651931763, "eval_runtime": 55.9536, "eval_samples_per_second": 9.061, "eval_steps_per_second": 1.144, "epoch": 4.0, "step": 4106}, {"loss": 1.3489, "grad_norm": 0.7726166248321533, "learning_rate": 0.0002, "epoch": 4.003896736483195, "step": 4110}, {"loss": 1.1415, "grad_norm": 1.1338967084884644, "learning_rate": 0.0002, "epoch": 4.013638577691184, "step": 4120}, {"loss": 1.2212, "grad_norm": 0.9530029296875, "learning_rate": 0.0002, "epoch": 4.023380418899172, "step": 4130}, {"loss": 1.2002, "grad_norm": 1.1058554649353027, "learning_rate": 0.0002, "epoch": 4.03312226010716, "step": 4140}, {"loss": 1.2381, "grad_norm": 0.8765049576759338, "learning_rate": 0.0002, "epoch": 4.042864101315149, "step": 4150}, {"loss": 1.2708, "grad_norm": 1.1774667501449585, "learning_rate": 0.0002, "epoch": 4.052605942523137, "step": 4160}, {"loss": 1.2116, "grad_norm": 0.9301433563232422, "learning_rate": 0.0002, "epoch": 4.062347783731125, "step": 4170}, {"loss": 1.1807, "grad_norm": 1.0196778774261475, "learning_rate": 0.0002, "epoch": 4.072089624939114, "step": 4180}, {"loss": 1.2602, "grad_norm": 1.1380577087402344, "learning_rate": 0.0002, "epoch": 4.081831466147102, "step": 4190}, {"loss": 1.2521, "grad_norm": 0.9121319651603699, "learning_rate": 0.0002, "epoch": 4.09157330735509, "step": 4200}, {"loss": 1.1747, "grad_norm": 0.9495378732681274, "learning_rate": 0.0002, "epoch": 4.101315148563079, "step": 4210}, {"loss": 1.1829, "grad_norm": 0.8058680295944214, "learning_rate": 0.0002, "epoch": 4.1110569897710665, "step": 4220}, {"loss": 1.1732, "grad_norm": 1.000887393951416, "learning_rate": 0.0002, "epoch": 4.120798830979055, "step": 4230}, {"loss": 1.1947, "grad_norm": 0.9529102444648743, "learning_rate": 0.0002, "epoch": 4.130540672187044, "step": 4240}, {"loss": 1.2104, "grad_norm": 1.0257115364074707, "learning_rate": 0.0002, "epoch": 4.140282513395031, "step": 4250}, {"loss": 1.2293, "grad_norm": 0.9590303897857666, "learning_rate": 0.0002, "epoch": 4.15002435460302, "step": 4260}, {"loss": 1.1918, "grad_norm": 1.065291166305542, "learning_rate": 0.0002, "epoch": 4.159766195811009, "step": 4270}, {"loss": 1.2323, "grad_norm": 0.8819697499275208, "learning_rate": 0.0002, "epoch": 4.169508037018996, "step": 4280}, {"loss": 1.2167, "grad_norm": 1.0335261821746826, "learning_rate": 0.0002, "epoch": 4.179249878226985, "step": 4290}, {"loss": 1.2131, "grad_norm": 0.8872809410095215, "learning_rate": 0.0002, "epoch": 4.1889917194349735, "step": 4300}, {"loss": 1.2794, "grad_norm": 0.9883159399032593, "learning_rate": 0.0002, "epoch": 4.198733560642961, "step": 4310}, {"loss": 1.2544, "grad_norm": 1.0254192352294922, "learning_rate": 0.0002, "epoch": 4.20847540185095, "step": 4320}, {"loss": 1.2595, "grad_norm": 0.9432600736618042, "learning_rate": 0.0002, "epoch": 4.218217243058938, "step": 4330}, {"loss": 1.2684, "grad_norm": 1.1008676290512085, "learning_rate": 0.0002, "epoch": 4.227959084266926, "step": 4340}, {"loss": 1.2149, "grad_norm": 1.0829699039459229, "learning_rate": 0.0002, "epoch": 4.237700925474915, "step": 4350}, {"loss": 1.2621, "grad_norm": 1.016847848892212, "learning_rate": 0.0002, "epoch": 4.247442766682903, "step": 4360}, {"loss": 1.2375, "grad_norm": 0.8924864530563354, "learning_rate": 0.0002, "epoch": 4.257184607890891, "step": 4370}, {"loss": 1.1987, "grad_norm": 0.9300530552864075, "learning_rate": 0.0002, "epoch": 4.26692644909888, "step": 4380}, {"loss": 1.1696, "grad_norm": 0.9684814810752869, "learning_rate": 0.0002, "epoch": 4.276668290306868, "step": 4390}, {"loss": 1.2006, "grad_norm": 0.9916250705718994, "learning_rate": 0.0002, "epoch": 4.286410131514856, "step": 4400}, {"loss": 1.2402, "grad_norm": 0.903680145740509, "learning_rate": 0.0002, "epoch": 4.2961519727228445, "step": 4410}, {"loss": 1.2022, "grad_norm": 0.8713505268096924, "learning_rate": 0.0002, "epoch": 4.305893813930833, "step": 4420}, {"loss": 1.1957, "grad_norm": 0.9983905553817749, "learning_rate": 0.0002, "epoch": 4.315635655138821, "step": 4430}, {"loss": 1.2676, "grad_norm": 1.1689040660858154, "learning_rate": 0.0002, "epoch": 4.3253774963468095, "step": 4440}, {"loss": 1.2166, "grad_norm": 0.9316853880882263, "learning_rate": 0.0002, "epoch": 4.335119337554798, "step": 4450}, {"loss": 1.222, "grad_norm": 0.9175887107849121, "learning_rate": 0.0002, "epoch": 4.344861178762786, "step": 4460}, {"loss": 1.2571, "grad_norm": 0.9348906874656677, "learning_rate": 0.0002, "epoch": 4.354603019970774, "step": 4470}, {"loss": 1.2764, "grad_norm": 0.9727016687393188, "learning_rate": 0.0002, "epoch": 4.364344861178763, "step": 4480}, {"loss": 1.2616, "grad_norm": 0.9843429923057556, "learning_rate": 0.0002, "epoch": 4.374086702386751, "step": 4490}, {"loss": 1.2488, "grad_norm": 0.9615852236747742, "learning_rate": 0.0002, "epoch": 4.383828543594739, "step": 4500}, {"loss": 1.1718, "grad_norm": 0.9688583612442017, "learning_rate": 0.0002, "epoch": 4.393570384802728, "step": 4510}, {"loss": 1.2546, "grad_norm": 0.9933668375015259, "learning_rate": 0.0002, "epoch": 4.403312226010716, "step": 4520}, {"loss": 1.2355, "grad_norm": 1.0626686811447144, "learning_rate": 0.0002, "epoch": 4.413054067218704, "step": 4530}, {"loss": 1.2425, "grad_norm": 0.9536267518997192, "learning_rate": 0.0002, "epoch": 4.422795908426693, "step": 4540}, {"loss": 1.2562, "grad_norm": 0.9777140021324158, "learning_rate": 0.0002, "epoch": 4.432537749634681, "step": 4550}, {"loss": 1.2878, "grad_norm": 0.980780839920044, "learning_rate": 0.0002, "epoch": 4.442279590842669, "step": 4560}, {"loss": 1.2597, "grad_norm": 1.0147196054458618, "learning_rate": 0.0002, "epoch": 4.452021432050658, "step": 4570}, {"loss": 1.2148, "grad_norm": 0.9763361811637878, "learning_rate": 0.0002, "epoch": 4.461763273258645, "step": 4580}, {"loss": 1.3076, "grad_norm": 1.0300798416137695, "learning_rate": 0.0002, "epoch": 4.471505114466634, "step": 4590}, {"loss": 1.2665, "grad_norm": 0.8833121657371521, "learning_rate": 0.0002, "epoch": 4.481246955674623, "step": 4600}, {"loss": 1.1899, "grad_norm": 1.1214020252227783, "learning_rate": 0.0002, "epoch": 4.490988796882611, "step": 4610}, {"loss": 1.2579, "grad_norm": 0.8843787908554077, "learning_rate": 0.0002, "epoch": 4.500730638090599, "step": 4620}, {"loss": 1.2633, "grad_norm": 0.9942020773887634, "learning_rate": 0.0002, "epoch": 4.5104724792985875, "step": 4630}, {"loss": 1.3172, "grad_norm": 1.0033202171325684, "learning_rate": 0.0002, "epoch": 4.520214320506576, "step": 4640}, {"loss": 1.2024, "grad_norm": 0.8767235279083252, "learning_rate": 0.0002, "epoch": 4.529956161714564, "step": 4650}, {"loss": 1.2714, "grad_norm": 1.0117276906967163, "learning_rate": 0.0002, "epoch": 4.539698002922552, "step": 4660}, {"loss": 1.2911, "grad_norm": 1.2787362337112427, "learning_rate": 0.0002, "epoch": 4.549439844130541, "step": 4670}, {"loss": 1.2603, "grad_norm": 0.8824878931045532, "learning_rate": 0.0002, "epoch": 4.559181685338529, "step": 4680}, {"loss": 1.2905, "grad_norm": 0.9209560751914978, "learning_rate": 0.0002, "epoch": 4.568923526546517, "step": 4690}, {"loss": 1.1916, "grad_norm": 1.1064010858535767, "learning_rate": 0.0002, "epoch": 4.578665367754506, "step": 4700}, {"loss": 1.2217, "grad_norm": 0.8914572596549988, "learning_rate": 0.0002, "epoch": 4.588407208962494, "step": 4710}, {"loss": 1.2861, "grad_norm": 1.0412265062332153, "learning_rate": 0.0002, "epoch": 4.598149050170482, "step": 4720}, {"loss": 1.262, "grad_norm": 1.1950221061706543, "learning_rate": 0.0002, "epoch": 4.607890891378471, "step": 4730}, {"loss": 1.2659, "grad_norm": 0.8938062787055969, "learning_rate": 0.0002, "epoch": 4.617632732586459, "step": 4740}, {"loss": 1.2621, "grad_norm": 0.9849569201469421, "learning_rate": 0.0002, "epoch": 4.627374573794447, "step": 4750}, {"loss": 1.2341, "grad_norm": 1.0081515312194824, "learning_rate": 0.0002, "epoch": 4.637116415002436, "step": 4760}, {"loss": 1.2023, "grad_norm": 0.8566309213638306, "learning_rate": 0.0002, "epoch": 4.6468582562104235, "step": 4770}, {"loss": 1.2723, "grad_norm": 1.1750118732452393, "learning_rate": 0.0002, "epoch": 4.656600097418412, "step": 4780}, {"loss": 1.2537, "grad_norm": 0.925502598285675, "learning_rate": 0.0002, "epoch": 4.666341938626401, "step": 4790}, {"loss": 1.2146, "grad_norm": 1.0402472019195557, "learning_rate": 0.0002, "epoch": 4.676083779834388, "step": 4800}, {"loss": 1.2555, "grad_norm": 0.9772472977638245, "learning_rate": 0.0002, "epoch": 4.685825621042377, "step": 4810}, {"loss": 1.2667, "grad_norm": 0.9082779288291931, "learning_rate": 0.0002, "epoch": 4.695567462250366, "step": 4820}, {"loss": 1.2465, "grad_norm": 0.8026862740516663, "learning_rate": 0.0002, "epoch": 4.705309303458353, "step": 4830}, {"loss": 1.3369, "grad_norm": 1.1631089448928833, "learning_rate": 0.0002, "epoch": 4.715051144666342, "step": 4840}, {"loss": 1.261, "grad_norm": 0.9384787678718567, "learning_rate": 0.0002, "epoch": 4.7247929858743305, "step": 4850}, {"loss": 1.2588, "grad_norm": 1.2151581048965454, "learning_rate": 0.0002, "epoch": 4.734534827082318, "step": 4860}, {"loss": 1.363, "grad_norm": 0.9679436087608337, "learning_rate": 0.0002, "epoch": 4.744276668290307, "step": 4870}, {"loss": 1.3292, "grad_norm": 0.8352158069610596, "learning_rate": 0.0002, "epoch": 4.754018509498295, "step": 4880}, {"loss": 1.3056, "grad_norm": 1.0205804109573364, "learning_rate": 0.0002, "epoch": 4.763760350706283, "step": 4890}, {"loss": 1.223, "grad_norm": 0.9814772605895996, "learning_rate": 0.0002, "epoch": 4.773502191914272, "step": 4900}, {"loss": 1.3114, "grad_norm": 1.002854347229004, "learning_rate": 0.0002, "epoch": 4.78324403312226, "step": 4910}, {"loss": 1.3143, "grad_norm": 1.1609505414962769, "learning_rate": 0.0002, "epoch": 4.792985874330248, "step": 4920}, {"loss": 1.3166, "grad_norm": 0.9354982376098633, "learning_rate": 0.0002, "epoch": 4.802727715538237, "step": 4930}, {"loss": 1.2978, "grad_norm": 0.9761685729026794, "learning_rate": 0.0002, "epoch": 4.812469556746225, "step": 4940}, {"loss": 1.2709, "grad_norm": 1.0604596138000488, "learning_rate": 0.0002, "epoch": 4.822211397954213, "step": 4950}, {"loss": 1.2765, "grad_norm": 1.0902808904647827, "learning_rate": 0.0002, "epoch": 4.8319532391622015, "step": 4960}, {"loss": 1.3073, "grad_norm": 1.0174955129623413, "learning_rate": 0.0002, "epoch": 4.84169508037019, "step": 4970}, {"loss": 1.3141, "grad_norm": 1.0995253324508667, "learning_rate": 0.0002, "epoch": 4.851436921578179, "step": 4980}, {"loss": 1.3006, "grad_norm": 0.880993127822876, "learning_rate": 0.0002, "epoch": 4.8611787627861665, "step": 4990}, {"loss": 1.2547, "grad_norm": 0.9472237825393677, "learning_rate": 0.0002, "epoch": 4.870920603994155, "step": 5000}, {"loss": 1.4078, "grad_norm": 0.9504236578941345, "learning_rate": 0.0002, "epoch": 4.880662445202143, "step": 5010}, {"loss": 1.2791, "grad_norm": 1.1261742115020752, "learning_rate": 0.0002, "epoch": 4.890404286410131, "step": 5020}, {"loss": 1.3707, "grad_norm": 0.904674768447876, "learning_rate": 0.0002, "epoch": 4.90014612761812, "step": 5030}, {"loss": 1.2762, "grad_norm": 0.8828991055488586, "learning_rate": 0.0002, "epoch": 4.909887968826109, "step": 5040}, {"loss": 1.2905, "grad_norm": 1.0156532526016235, "learning_rate": 0.0002, "epoch": 4.919629810034096, "step": 5050}, {"loss": 1.3079, "grad_norm": 0.8975168466567993, "learning_rate": 0.0002, "epoch": 4.929371651242085, "step": 5060}, {"loss": 1.3322, "grad_norm": 0.9787213802337646, "learning_rate": 0.0002, "epoch": 4.939113492450073, "step": 5070}, {"loss": 1.2533, "grad_norm": 1.0801568031311035, "learning_rate": 0.0002, "epoch": 4.948855333658061, "step": 5080}, {"loss": 1.238, "grad_norm": 1.0655089616775513, "learning_rate": 0.0002, "epoch": 4.95859717486605, "step": 5090}, {"loss": 1.2449, "grad_norm": 0.8941320180892944, "learning_rate": 0.0002, "epoch": 4.968339016074038, "step": 5100}, {"loss": 1.2846, "grad_norm": 1.050621747970581, "learning_rate": 0.0002, "epoch": 4.978080857282026, "step": 5110}, {"loss": 1.3791, "grad_norm": 0.9724781513214111, "learning_rate": 0.0002, "epoch": 4.987822698490015, "step": 5120}, {"loss": 1.292, "grad_norm": 0.9850538969039917, "learning_rate": 0.0002, "epoch": 4.997564539698003, "step": 5130}, {"eval_loss": 2.0824170112609863, "eval_runtime": 55.592, "eval_samples_per_second": 9.12, "eval_steps_per_second": 1.151, "epoch": 4.9995129079396, "step": 5132}, {"loss": 1.037, "grad_norm": 1.0096189975738525, "learning_rate": 0.0002, "epoch": 5.007306380905991, "step": 5140}, {"loss": 1.0003, "grad_norm": 1.2403408288955688, "learning_rate": 0.0002, "epoch": 5.01704822211398, "step": 5150}, {"loss": 1.0129, "grad_norm": 1.1243221759796143, "learning_rate": 0.0002, "epoch": 5.026790063321968, "step": 5160}, {"loss": 0.9815, "grad_norm": 1.4745502471923828, "learning_rate": 0.0002, "epoch": 5.036531904529956, "step": 5170}, {"loss": 0.9715, "grad_norm": 1.1913198232650757, "learning_rate": 0.0002, "epoch": 5.0462737457379445, "step": 5180}, {"loss": 0.9282, "grad_norm": 1.2732855081558228, "learning_rate": 0.0002, "epoch": 5.056015586945933, "step": 5190}, {"loss": 0.9857, "grad_norm": 1.1737396717071533, "learning_rate": 0.0002, "epoch": 5.065757428153921, "step": 5200}, {"loss": 0.9754, "grad_norm": 1.4162768125534058, "learning_rate": 0.0002, "epoch": 5.075499269361909, "step": 5210}, {"loss": 1.0333, "grad_norm": 1.528274655342102, "learning_rate": 0.0002, "epoch": 5.085241110569898, "step": 5220}, {"loss": 1.0227, "grad_norm": 1.3966618776321411, "learning_rate": 0.0002, "epoch": 5.094982951777886, "step": 5230}, {"loss": 0.987, "grad_norm": 1.3427953720092773, "learning_rate": 0.0002, "epoch": 5.104724792985874, "step": 5240}, {"loss": 1.0353, "grad_norm": 1.6533905267715454, "learning_rate": 0.0002, "epoch": 5.114466634193863, "step": 5250}, {"loss": 1.0452, "grad_norm": 1.4114865064620972, "learning_rate": 0.0002, "epoch": 5.124208475401851, "step": 5260}, {"loss": 1.067, "grad_norm": 1.5460708141326904, "learning_rate": 0.0002, "epoch": 5.133950316609839, "step": 5270}, {"loss": 1.0667, "grad_norm": 1.3491919040679932, "learning_rate": 0.0002, "epoch": 5.143692157817828, "step": 5280}, {"loss": 0.9957, "grad_norm": 1.2208969593048096, "learning_rate": 0.0002, "epoch": 5.153433999025816, "step": 5290}, {"loss": 1.0362, "grad_norm": 1.1141403913497925, "learning_rate": 0.0002, "epoch": 5.163175840233804, "step": 5300}, {"loss": 0.9744, "grad_norm": 1.2938064336776733, "learning_rate": 0.0002, "epoch": 5.172917681441793, "step": 5310}, {"loss": 1.0438, "grad_norm": 1.2704918384552002, "learning_rate": 0.0002, "epoch": 5.1826595226497805, "step": 5320}, {"loss": 1.0015, "grad_norm": 1.3928544521331787, "learning_rate": 0.0002, "epoch": 5.192401363857769, "step": 5330}, {"loss": 1.025, "grad_norm": 1.1993824243545532, "learning_rate": 0.0002, "epoch": 5.202143205065758, "step": 5340}, {"loss": 1.0195, "grad_norm": 1.5913670063018799, "learning_rate": 0.0002, "epoch": 5.211885046273745, "step": 5350}, {"loss": 1.0113, "grad_norm": 1.1577855348587036, "learning_rate": 0.0002, "epoch": 5.221626887481734, "step": 5360}, {"loss": 1.0684, "grad_norm": 1.4535993337631226, "learning_rate": 0.0002, "epoch": 5.231368728689723, "step": 5370}, {"loss": 1.0255, "grad_norm": 1.5068976879119873, "learning_rate": 0.0002, "epoch": 5.24111056989771, "step": 5380}, {"loss": 1.0068, "grad_norm": 1.2365459203720093, "learning_rate": 0.0002, "epoch": 5.250852411105699, "step": 5390}, {"loss": 1.0145, "grad_norm": 1.3197922706604004, "learning_rate": 0.0002, "epoch": 5.2605942523136875, "step": 5400}, {"loss": 1.0767, "grad_norm": 1.2395117282867432, "learning_rate": 0.0002, "epoch": 5.270336093521675, "step": 5410}, {"loss": 1.0292, "grad_norm": 1.1841236352920532, "learning_rate": 0.0002, "epoch": 5.280077934729664, "step": 5420}, {"loss": 1.0233, "grad_norm": 1.218003749847412, "learning_rate": 0.0002, "epoch": 5.289819775937652, "step": 5430}, {"loss": 1.0093, "grad_norm": 1.2210947275161743, "learning_rate": 0.0002, "epoch": 5.29956161714564, "step": 5440}, {"loss": 0.9619, "grad_norm": 1.266006588935852, "learning_rate": 0.0002, "epoch": 5.309303458353629, "step": 5450}, {"loss": 1.0352, "grad_norm": 1.2598075866699219, "learning_rate": 0.0002, "epoch": 5.319045299561617, "step": 5460}, {"loss": 1.0929, "grad_norm": 1.2410019636154175, "learning_rate": 0.0002, "epoch": 5.328787140769606, "step": 5470}, {"loss": 1.058, "grad_norm": 1.249698519706726, "learning_rate": 0.0002, "epoch": 5.338528981977594, "step": 5480}, {"loss": 1.0457, "grad_norm": 1.2398173809051514, "learning_rate": 0.0002, "epoch": 5.348270823185582, "step": 5490}, {"loss": 1.0139, "grad_norm": 1.2416654825210571, "learning_rate": 0.0002, "epoch": 5.35801266439357, "step": 5500}, {"loss": 1.0609, "grad_norm": 1.398706316947937, "learning_rate": 0.0002, "epoch": 5.3677545056015585, "step": 5510}, {"loss": 1.0512, "grad_norm": 1.3049418926239014, "learning_rate": 0.0002, "epoch": 5.377496346809547, "step": 5520}, {"loss": 1.0912, "grad_norm": 1.2528893947601318, "learning_rate": 0.0002, "epoch": 5.387238188017536, "step": 5530}, {"loss": 1.0619, "grad_norm": 1.2963255643844604, "learning_rate": 0.0002, "epoch": 5.3969800292255234, "step": 5540}, {"loss": 1.0194, "grad_norm": 1.494231104850769, "learning_rate": 0.0002, "epoch": 5.406721870433512, "step": 5550}, {"loss": 1.0179, "grad_norm": 1.2760992050170898, "learning_rate": 0.0002, "epoch": 5.416463711641501, "step": 5560}, {"loss": 1.1088, "grad_norm": 1.195292592048645, "learning_rate": 0.0002, "epoch": 5.426205552849488, "step": 5570}, {"loss": 1.0859, "grad_norm": 1.6408965587615967, "learning_rate": 0.0002, "epoch": 5.435947394057477, "step": 5580}, {"loss": 1.0868, "grad_norm": 1.3092058897018433, "learning_rate": 0.0002, "epoch": 5.4456892352654656, "step": 5590}, {"loss": 1.006, "grad_norm": 1.2960586547851562, "learning_rate": 0.0002, "epoch": 5.455431076473453, "step": 5600}, {"loss": 1.0257, "grad_norm": 1.3560487031936646, "learning_rate": 0.0002, "epoch": 5.465172917681442, "step": 5610}, {"loss": 1.0314, "grad_norm": 1.1896311044692993, "learning_rate": 0.0002, "epoch": 5.4749147588894305, "step": 5620}, {"loss": 1.0435, "grad_norm": 1.3145595788955688, "learning_rate": 0.0002, "epoch": 5.484656600097418, "step": 5630}, {"loss": 1.0456, "grad_norm": 1.2207404375076294, "learning_rate": 0.0002, "epoch": 5.494398441305407, "step": 5640}, {"loss": 1.0823, "grad_norm": 1.266015887260437, "learning_rate": 0.0002, "epoch": 5.504140282513395, "step": 5650}, {"loss": 1.0696, "grad_norm": 1.2478289604187012, "learning_rate": 0.0002, "epoch": 5.513882123721383, "step": 5660}, {"loss": 1.0695, "grad_norm": 1.4851372241973877, "learning_rate": 0.0002, "epoch": 5.523623964929372, "step": 5670}, {"loss": 1.0736, "grad_norm": 1.4478679895401, "learning_rate": 0.0002, "epoch": 5.53336580613736, "step": 5680}, {"loss": 1.043, "grad_norm": 1.1079537868499756, "learning_rate": 0.0002, "epoch": 5.543107647345348, "step": 5690}, {"loss": 1.1107, "grad_norm": 1.4201879501342773, "learning_rate": 0.0002, "epoch": 5.552849488553337, "step": 5700}, {"loss": 1.0697, "grad_norm": 1.2092000246047974, "learning_rate": 0.0002, "epoch": 5.562591329761325, "step": 5710}, {"loss": 0.9868, "grad_norm": 1.4515851736068726, "learning_rate": 0.0002, "epoch": 5.572333170969313, "step": 5720}, {"loss": 1.1547, "grad_norm": 1.3260412216186523, "learning_rate": 0.0002, "epoch": 5.5820750121773015, "step": 5730}, {"loss": 1.1388, "grad_norm": 1.248191475868225, "learning_rate": 0.0002, "epoch": 5.59181685338529, "step": 5740}, {"loss": 1.0597, "grad_norm": 1.2037307024002075, "learning_rate": 0.0002, "epoch": 5.601558694593278, "step": 5750}, {"loss": 1.1425, "grad_norm": 1.341237187385559, "learning_rate": 0.0002, "epoch": 5.611300535801266, "step": 5760}, {"loss": 1.0942, "grad_norm": 1.130115270614624, "learning_rate": 0.0002, "epoch": 5.621042377009255, "step": 5770}, {"loss": 1.1029, "grad_norm": 1.3834772109985352, "learning_rate": 0.0002, "epoch": 5.630784218217243, "step": 5780}, {"loss": 1.0825, "grad_norm": 1.2586270570755005, "learning_rate": 0.0002, "epoch": 5.640526059425231, "step": 5790}, {"loss": 1.0186, "grad_norm": 1.3233023881912231, "learning_rate": 0.0002, "epoch": 5.65026790063322, "step": 5800}, {"loss": 1.0557, "grad_norm": 1.2711341381072998, "learning_rate": 0.0002, "epoch": 5.660009741841208, "step": 5810}, {"loss": 1.0897, "grad_norm": 1.3867720365524292, "learning_rate": 0.0002, "epoch": 5.669751583049196, "step": 5820}, {"loss": 1.0776, "grad_norm": 1.4783269166946411, "learning_rate": 0.0002, "epoch": 5.679493424257185, "step": 5830}, {"loss": 1.0632, "grad_norm": 1.2744768857955933, "learning_rate": 0.0002, "epoch": 5.6892352654651726, "step": 5840}, {"loss": 1.1484, "grad_norm": 1.3405882120132446, "learning_rate": 0.0002, "epoch": 5.698977106673161, "step": 5850}, {"loss": 1.0975, "grad_norm": 1.204300880432129, "learning_rate": 0.0002, "epoch": 5.70871894788115, "step": 5860}, {"loss": 1.0494, "grad_norm": 1.2954572439193726, "learning_rate": 0.0002, "epoch": 5.7184607890891375, "step": 5870}, {"loss": 1.0643, "grad_norm": 1.5478382110595703, "learning_rate": 0.0002, "epoch": 5.728202630297126, "step": 5880}, {"loss": 1.0582, "grad_norm": 1.2095842361450195, "learning_rate": 0.0002, "epoch": 5.737944471505115, "step": 5890}, {"loss": 1.1, "grad_norm": 1.0691519975662231, "learning_rate": 0.0002, "epoch": 5.747686312713103, "step": 5900}, {"loss": 1.0906, "grad_norm": 1.1920677423477173, "learning_rate": 0.0002, "epoch": 5.757428153921091, "step": 5910}, {"loss": 1.1746, "grad_norm": 1.2051277160644531, "learning_rate": 0.0002, "epoch": 5.76716999512908, "step": 5920}, {"loss": 1.1221, "grad_norm": 1.197490930557251, "learning_rate": 0.0002, "epoch": 5.776911836337067, "step": 5930}, {"loss": 1.07, "grad_norm": 1.2003998756408691, "learning_rate": 0.0002, "epoch": 5.786653677545056, "step": 5940}, {"loss": 1.0938, "grad_norm": 1.2323646545410156, "learning_rate": 0.0002, "epoch": 5.7963955187530445, "step": 5950}, {"loss": 1.1443, "grad_norm": 1.2593932151794434, "learning_rate": 0.0002, "epoch": 5.806137359961033, "step": 5960}, {"loss": 1.0829, "grad_norm": 1.1835976839065552, "learning_rate": 0.0002, "epoch": 5.815879201169021, "step": 5970}, {"loss": 1.1056, "grad_norm": 1.4770104885101318, "learning_rate": 0.0002, "epoch": 5.825621042377009, "step": 5980}, {"loss": 1.1934, "grad_norm": 1.1025809049606323, "learning_rate": 0.0002, "epoch": 5.835362883584997, "step": 5990}, {"loss": 1.1323, "grad_norm": 1.364588975906372, "learning_rate": 0.0002, "epoch": 5.845104724792986, "step": 6000}, {"loss": 1.1234, "grad_norm": 1.2340112924575806, "learning_rate": 0.0002, "epoch": 5.854846566000974, "step": 6010}, {"loss": 1.1123, "grad_norm": 1.4925711154937744, "learning_rate": 0.0002, "epoch": 5.864588407208963, "step": 6020}, {"loss": 1.12, "grad_norm": 1.3516744375228882, "learning_rate": 0.0002, "epoch": 5.874330248416951, "step": 6030}, {"loss": 1.1399, "grad_norm": 1.2058138847351074, "learning_rate": 0.0002, "epoch": 5.884072089624939, "step": 6040}, {"loss": 1.1074, "grad_norm": 1.13870108127594, "learning_rate": 0.0002, "epoch": 5.893813930832927, "step": 6050}, {"loss": 1.088, "grad_norm": 1.1587319374084473, "learning_rate": 0.0002, "epoch": 5.9035557720409155, "step": 6060}, {"loss": 1.1376, "grad_norm": 1.164481520652771, "learning_rate": 0.0002, "epoch": 5.913297613248904, "step": 6070}, {"loss": 1.1262, "grad_norm": 1.2115206718444824, "learning_rate": 0.0002, "epoch": 5.923039454456893, "step": 6080}, {"loss": 1.1345, "grad_norm": 1.3201590776443481, "learning_rate": 0.0002, "epoch": 5.93278129566488, "step": 6090}, {"loss": 1.1288, "grad_norm": 1.287380576133728, "learning_rate": 0.0002, "epoch": 5.942523136872869, "step": 6100}, {"loss": 1.1475, "grad_norm": 1.1820166110992432, "learning_rate": 0.0002, "epoch": 5.952264978080858, "step": 6110}, {"loss": 1.1112, "grad_norm": 1.2550667524337769, "learning_rate": 0.0002, "epoch": 5.962006819288845, "step": 6120}, {"loss": 1.1528, "grad_norm": 1.3547813892364502, "learning_rate": 0.0002, "epoch": 5.971748660496834, "step": 6130}, {"loss": 1.0557, "grad_norm": 1.260842204093933, "learning_rate": 0.0002, "epoch": 5.9814905017048225, "step": 6140}, {"loss": 1.1119, "grad_norm": 1.1643036603927612, "learning_rate": 0.0002, "epoch": 5.99123234291281, "step": 6150}, {"eval_loss": 2.2628161907196045, "eval_runtime": 57.2379, "eval_samples_per_second": 8.858, "eval_steps_per_second": 1.118, "epoch": 6.0, "step": 6159}, {"loss": 1.0837, "grad_norm": 0.9384723901748657, "learning_rate": 0.0002, "epoch": 6.000974184120799, "step": 6160}, {"loss": 0.7335, "grad_norm": 2.1525821685791016, "learning_rate": 0.0002, "epoch": 6.0107160253287875, "step": 6170}, {"loss": 0.8416, "grad_norm": 2.0194077491760254, "learning_rate": 0.0002, "epoch": 6.020457866536775, "step": 6180}, {"loss": 0.8443, "grad_norm": 1.5257816314697266, "learning_rate": 0.0002, "epoch": 6.030199707744764, "step": 6190}, {"loss": 0.7543, "grad_norm": 1.5432662963867188, "learning_rate": 0.0002, "epoch": 6.039941548952752, "step": 6200}, {"loss": 0.8104, "grad_norm": 1.6874405145645142, "learning_rate": 0.0002, "epoch": 6.04968339016074, "step": 6210}, {"loss": 0.8395, "grad_norm": 1.7346407175064087, "learning_rate": 0.0002, "epoch": 6.059425231368729, "step": 6220}, {"loss": 0.8027, "grad_norm": 1.5320781469345093, "learning_rate": 0.0002, "epoch": 6.069167072576717, "step": 6230}, {"loss": 0.7488, "grad_norm": 1.4106669425964355, "learning_rate": 0.0002, "epoch": 6.078908913784705, "step": 6240}, {"loss": 0.812, "grad_norm": 1.5568628311157227, "learning_rate": 0.0002, "epoch": 6.088650754992694, "step": 6250}, {"loss": 0.8055, "grad_norm": 1.6155978441238403, "learning_rate": 0.0002, "epoch": 6.098392596200682, "step": 6260}, {"loss": 0.8225, "grad_norm": 1.4820445775985718, "learning_rate": 0.0002, "epoch": 6.10813443740867, "step": 6270}, {"loss": 0.8599, "grad_norm": 1.6163820028305054, "learning_rate": 0.0002, "epoch": 6.1178762786166585, "step": 6280}, {"loss": 0.853, "grad_norm": 1.8396387100219727, "learning_rate": 0.0002, "epoch": 6.127618119824647, "step": 6290}, {"loss": 0.7768, "grad_norm": 1.7181230783462524, "learning_rate": 0.0002, "epoch": 6.137359961032635, "step": 6300}, {"loss": 0.8116, "grad_norm": 1.6568509340286255, "learning_rate": 0.0002, "epoch": 6.147101802240623, "step": 6310}, {"loss": 0.8525, "grad_norm": 1.3481947183609009, "learning_rate": 0.0002, "epoch": 6.156843643448612, "step": 6320}, {"loss": 0.762, "grad_norm": 1.5788342952728271, "learning_rate": 0.0002, "epoch": 6.1665854846566, "step": 6330}, {"loss": 0.886, "grad_norm": 1.5067620277404785, "learning_rate": 0.0002, "epoch": 6.176327325864588, "step": 6340}, {"loss": 0.8375, "grad_norm": 1.8198208808898926, "learning_rate": 0.0002, "epoch": 6.186069167072577, "step": 6350}, {"loss": 0.7867, "grad_norm": 1.4012749195098877, "learning_rate": 0.0002, "epoch": 6.195811008280565, "step": 6360}, {"loss": 0.8144, "grad_norm": 1.759798288345337, "learning_rate": 0.0002, "epoch": 6.205552849488553, "step": 6370}, {"loss": 0.7811, "grad_norm": 1.468922734260559, "learning_rate": 0.0002, "epoch": 6.215294690696542, "step": 6380}, {"loss": 0.8356, "grad_norm": 1.3706471920013428, "learning_rate": 0.0002, "epoch": 6.2250365319045295, "step": 6390}, {"loss": 0.8096, "grad_norm": 1.6397383213043213, "learning_rate": 0.0002, "epoch": 6.234778373112518, "step": 6400}, {"loss": 0.8834, "grad_norm": 1.5614187717437744, "learning_rate": 0.0002, "epoch": 6.244520214320507, "step": 6410}, {"loss": 0.8533, "grad_norm": 1.7118678092956543, "learning_rate": 0.0002, "epoch": 6.2542620555284945, "step": 6420}, {"loss": 0.8653, "grad_norm": 1.4041547775268555, "learning_rate": 0.0002, "epoch": 6.264003896736483, "step": 6430}, {"loss": 0.879, "grad_norm": 1.7653605937957764, "learning_rate": 0.0002, "epoch": 6.273745737944472, "step": 6440}, {"loss": 0.8786, "grad_norm": 2.6219191551208496, "learning_rate": 0.0002, "epoch": 6.28348757915246, "step": 6450}, {"loss": 0.8896, "grad_norm": 1.4757837057113647, "learning_rate": 0.0002, "epoch": 6.293229420360448, "step": 6460}, {"loss": 0.9079, "grad_norm": 1.715598225593567, "learning_rate": 0.0002, "epoch": 6.302971261568437, "step": 6470}, {"loss": 0.8526, "grad_norm": 1.376216173171997, "learning_rate": 0.0002, "epoch": 6.312713102776424, "step": 6480}, {"loss": 0.8742, "grad_norm": 1.7119828462600708, "learning_rate": 0.0002, "epoch": 6.322454943984413, "step": 6490}, {"loss": 0.7988, "grad_norm": 1.4304355382919312, "learning_rate": 0.0002, "epoch": 6.3321967851924015, "step": 6500}, {"loss": 0.8539, "grad_norm": 1.4889872074127197, "learning_rate": 0.0002, "epoch": 6.34193862640039, "step": 6510}, {"loss": 0.9328, "grad_norm": 1.370373010635376, "learning_rate": 0.0002, "epoch": 6.351680467608378, "step": 6520}, {"loss": 0.8997, "grad_norm": 1.7697709798812866, "learning_rate": 0.0002, "epoch": 6.361422308816366, "step": 6530}, {"loss": 0.9421, "grad_norm": 1.495297908782959, "learning_rate": 0.0002, "epoch": 6.371164150024355, "step": 6540}, {"loss": 0.8796, "grad_norm": 1.7251347303390503, "learning_rate": 0.0002, "epoch": 6.380905991232343, "step": 6550}, {"loss": 0.9327, "grad_norm": 1.6909505128860474, "learning_rate": 0.0002, "epoch": 6.390647832440331, "step": 6560}, {"loss": 0.837, "grad_norm": 1.4369314908981323, "learning_rate": 0.0002, "epoch": 6.40038967364832, "step": 6570}, {"loss": 0.8572, "grad_norm": 1.7803739309310913, "learning_rate": 0.0002, "epoch": 6.410131514856308, "step": 6580}, {"loss": 0.9024, "grad_norm": 1.6107097864151, "learning_rate": 0.0002, "epoch": 6.419873356064296, "step": 6590}, {"loss": 0.8469, "grad_norm": 1.6151643991470337, "learning_rate": 0.0002, "epoch": 6.429615197272285, "step": 6600}, {"loss": 0.8791, "grad_norm": 1.7159833908081055, "learning_rate": 0.0002, "epoch": 6.4393570384802725, "step": 6610}, {"loss": 0.9249, "grad_norm": 1.4366064071655273, "learning_rate": 0.0002, "epoch": 6.449098879688261, "step": 6620}, {"loss": 0.8417, "grad_norm": 1.6050453186035156, "learning_rate": 0.0002, "epoch": 6.45884072089625, "step": 6630}, {"loss": 0.8943, "grad_norm": 1.6296740770339966, "learning_rate": 0.0002, "epoch": 6.468582562104237, "step": 6640}, {"loss": 0.9228, "grad_norm": 1.6181174516677856, "learning_rate": 0.0002, "epoch": 6.478324403312226, "step": 6650}, {"loss": 0.9139, "grad_norm": 1.5452176332473755, "learning_rate": 0.0002, "epoch": 6.488066244520215, "step": 6660}, {"loss": 0.9022, "grad_norm": 1.3919731378555298, "learning_rate": 0.0002, "epoch": 6.497808085728202, "step": 6670}, {"loss": 0.9046, "grad_norm": 1.6456257104873657, "learning_rate": 0.0002, "epoch": 6.507549926936191, "step": 6680}, {"loss": 0.9041, "grad_norm": 1.4147369861602783, "learning_rate": 0.0002, "epoch": 6.5172917681441795, "step": 6690}, {"loss": 0.8361, "grad_norm": 1.7005025148391724, "learning_rate": 0.0002, "epoch": 6.527033609352167, "step": 6700}, {"loss": 0.8738, "grad_norm": 1.6032357215881348, "learning_rate": 0.0002, "epoch": 6.536775450560156, "step": 6710}, {"loss": 0.9796, "grad_norm": 1.3454229831695557, "learning_rate": 0.0002, "epoch": 6.5465172917681445, "step": 6720}, {"loss": 0.8573, "grad_norm": 1.6961418390274048, "learning_rate": 0.0002, "epoch": 6.556259132976132, "step": 6730}, {"loss": 0.9241, "grad_norm": 1.78407883644104, "learning_rate": 0.0002, "epoch": 6.566000974184121, "step": 6740}, {"loss": 0.8941, "grad_norm": 1.6817889213562012, "learning_rate": 0.0002, "epoch": 6.575742815392109, "step": 6750}, {"loss": 0.8765, "grad_norm": 1.7894943952560425, "learning_rate": 0.0002, "epoch": 6.585484656600097, "step": 6760}, {"loss": 0.8607, "grad_norm": 1.6404837369918823, "learning_rate": 0.0002, "epoch": 6.595226497808086, "step": 6770}, {"loss": 0.8573, "grad_norm": 1.5849255323410034, "learning_rate": 0.0002, "epoch": 6.604968339016074, "step": 6780}, {"loss": 0.9575, "grad_norm": 1.5993813276290894, "learning_rate": 0.0002, "epoch": 6.614710180224062, "step": 6790}, {"loss": 0.8922, "grad_norm": 1.2834863662719727, "learning_rate": 0.0002, "epoch": 6.624452021432051, "step": 6800}, {"loss": 0.9007, "grad_norm": 1.7215641736984253, "learning_rate": 0.0002, "epoch": 6.634193862640039, "step": 6810}, {"loss": 0.9292, "grad_norm": 1.7588146924972534, "learning_rate": 0.0002, "epoch": 6.643935703848027, "step": 6820}, {"loss": 0.8634, "grad_norm": 1.7956023216247559, "learning_rate": 0.0002, "epoch": 6.6536775450560155, "step": 6830}, {"loss": 0.8108, "grad_norm": 1.5115351676940918, "learning_rate": 0.0002, "epoch": 6.663419386264004, "step": 6840}, {"loss": 0.9329, "grad_norm": 1.5660319328308105, "learning_rate": 0.0002, "epoch": 6.673161227471992, "step": 6850}, {"loss": 0.9877, "grad_norm": 1.4323679208755493, "learning_rate": 0.0002, "epoch": 6.68290306867998, "step": 6860}, {"loss": 0.8732, "grad_norm": 1.662089467048645, "learning_rate": 0.0002, "epoch": 6.692644909887969, "step": 6870}, {"loss": 0.87, "grad_norm": 1.7854869365692139, "learning_rate": 0.0002, "epoch": 6.702386751095958, "step": 6880}, {"loss": 0.9105, "grad_norm": 1.5491222143173218, "learning_rate": 0.0002, "epoch": 6.712128592303945, "step": 6890}, {"loss": 0.9147, "grad_norm": 1.5946987867355347, "learning_rate": 0.0002, "epoch": 6.721870433511934, "step": 6900}, {"loss": 0.9391, "grad_norm": 1.6195964813232422, "learning_rate": 0.0002, "epoch": 6.731612274719922, "step": 6910}, {"loss": 0.8947, "grad_norm": 1.6366901397705078, "learning_rate": 0.0002, "epoch": 6.74135411592791, "step": 6920}, {"loss": 0.8695, "grad_norm": 1.5080382823944092, "learning_rate": 0.0002, "epoch": 6.751095957135899, "step": 6930}, {"loss": 0.9124, "grad_norm": 1.742353916168213, "learning_rate": 0.0002, "epoch": 6.760837798343887, "step": 6940}, {"loss": 0.9118, "grad_norm": 1.690251111984253, "learning_rate": 0.0002, "epoch": 6.770579639551875, "step": 6950}, {"loss": 0.9039, "grad_norm": 1.7103357315063477, "learning_rate": 0.0002, "epoch": 6.780321480759864, "step": 6960}, {"loss": 0.869, "grad_norm": 1.6630914211273193, "learning_rate": 0.0002, "epoch": 6.7900633219678515, "step": 6970}, {"loss": 0.8944, "grad_norm": 1.423768162727356, "learning_rate": 0.0002, "epoch": 6.79980516317584, "step": 6980}, {"loss": 0.9397, "grad_norm": 1.7844693660736084, "learning_rate": 0.0002, "epoch": 6.809547004383829, "step": 6990}, {"loss": 0.8889, "grad_norm": 1.545282006263733, "learning_rate": 0.0002, "epoch": 6.819288845591817, "step": 7000}, {"loss": 0.9333, "grad_norm": 1.4340319633483887, "learning_rate": 0.0002, "epoch": 6.829030686799805, "step": 7010}, {"loss": 0.9486, "grad_norm": 1.5981626510620117, "learning_rate": 0.0002, "epoch": 6.838772528007794, "step": 7020}, {"loss": 0.9062, "grad_norm": 1.5205026865005493, "learning_rate": 0.0002, "epoch": 6.848514369215782, "step": 7030}, {"loss": 0.9245, "grad_norm": 1.6999989748001099, "learning_rate": 0.0002, "epoch": 6.85825621042377, "step": 7040}, {"loss": 0.9313, "grad_norm": 1.6392347812652588, "learning_rate": 0.0002, "epoch": 6.8679980516317585, "step": 7050}, {"loss": 0.9275, "grad_norm": 1.637308955192566, "learning_rate": 0.0002, "epoch": 6.877739892839747, "step": 7060}, {"loss": 0.9672, "grad_norm": 1.671341896057129, "learning_rate": 0.0002, "epoch": 6.887481734047735, "step": 7070}, {"loss": 0.9726, "grad_norm": 1.4437555074691772, "learning_rate": 0.0002, "epoch": 6.897223575255723, "step": 7080}, {"loss": 0.9454, "grad_norm": 1.4251935482025146, "learning_rate": 0.0002, "epoch": 6.906965416463712, "step": 7090}, {"loss": 0.8858, "grad_norm": 1.5106734037399292, "learning_rate": 0.0002, "epoch": 6.9167072576717, "step": 7100}, {"loss": 0.939, "grad_norm": 1.670742154121399, "learning_rate": 0.0002, "epoch": 6.926449098879688, "step": 7110}, {"loss": 0.8818, "grad_norm": 1.4353723526000977, "learning_rate": 0.0002, "epoch": 6.936190940087677, "step": 7120}, {"loss": 0.9354, "grad_norm": 1.9437772035598755, "learning_rate": 0.0002, "epoch": 6.945932781295665, "step": 7130}, {"loss": 0.9623, "grad_norm": 1.4922038316726685, "learning_rate": 0.0002, "epoch": 6.955674622503653, "step": 7140}, {"loss": 0.9653, "grad_norm": 1.489193081855774, "learning_rate": 0.0002, "epoch": 6.965416463711642, "step": 7150}, {"loss": 1.0024, "grad_norm": 1.529490351676941, "learning_rate": 0.0002, "epoch": 6.9751583049196295, "step": 7160}, {"loss": 0.9715, "grad_norm": 1.7370105981826782, "learning_rate": 0.0002, "epoch": 6.984900146127618, "step": 7170}, {"loss": 0.921, "grad_norm": 1.5639604330062866, "learning_rate": 0.0002, "epoch": 6.994641987335607, "step": 7180}, {"eval_loss": 2.521758794784546, "eval_runtime": 56.1587, "eval_samples_per_second": 9.028, "eval_steps_per_second": 1.14, "epoch": 6.9995129079396, "step": 7185}, {"loss": 0.8682, "grad_norm": 1.391621470451355, "learning_rate": 0.0002, "epoch": 7.004383828543594, "step": 7190}, {"loss": 0.6591, "grad_norm": 2.3696491718292236, "learning_rate": 0.0002, "epoch": 7.014125669751583, "step": 7200}, {"loss": 0.5653, "grad_norm": 1.6873828172683716, "learning_rate": 0.0002, "epoch": 7.023867510959572, "step": 7210}, {"loss": 0.6301, "grad_norm": 1.8893300294876099, "learning_rate": 0.0002, "epoch": 7.033609352167559, "step": 7220}, {"loss": 0.5653, "grad_norm": 1.6323082447052002, "learning_rate": 0.0002, "epoch": 7.043351193375548, "step": 7230}, {"loss": 0.6402, "grad_norm": 1.9979127645492554, "learning_rate": 0.0002, "epoch": 7.0530930345835365, "step": 7240}, {"loss": 0.6314, "grad_norm": 2.0339183807373047, "learning_rate": 0.0002, "epoch": 7.062834875791524, "step": 7250}, {"loss": 0.615, "grad_norm": 1.6820781230926514, "learning_rate": 0.0002, "epoch": 7.072576716999513, "step": 7260}, {"loss": 0.7023, "grad_norm": 2.0400710105895996, "learning_rate": 0.0002, "epoch": 7.0823185582075014, "step": 7270}, {"loss": 0.6003, "grad_norm": 2.13495135307312, "learning_rate": 0.0002, "epoch": 7.092060399415489, "step": 7280}, {"loss": 0.6243, "grad_norm": 1.6993554830551147, "learning_rate": 0.0002, "epoch": 7.101802240623478, "step": 7290}, {"loss": 0.6587, "grad_norm": 1.9262464046478271, "learning_rate": 0.0002, "epoch": 7.111544081831466, "step": 7300}, {"loss": 0.6445, "grad_norm": 1.8407244682312012, "learning_rate": 0.0002, "epoch": 7.121285923039454, "step": 7310}, {"loss": 0.6305, "grad_norm": 1.744294285774231, "learning_rate": 0.0002, "epoch": 7.131027764247443, "step": 7320}, {"loss": 0.6886, "grad_norm": 1.7602320909500122, "learning_rate": 0.0002, "epoch": 7.140769605455431, "step": 7330}, {"loss": 0.6582, "grad_norm": 1.7360851764678955, "learning_rate": 0.0002, "epoch": 7.150511446663419, "step": 7340}, {"loss": 0.6172, "grad_norm": 2.0012850761413574, "learning_rate": 0.0002, "epoch": 7.160253287871408, "step": 7350}, {"loss": 0.6636, "grad_norm": 2.064319372177124, "learning_rate": 0.0002, "epoch": 7.169995129079396, "step": 7360}, {"loss": 0.6273, "grad_norm": 1.4556169509887695, "learning_rate": 0.0002, "epoch": 7.179736970287384, "step": 7370}, {"loss": 0.6944, "grad_norm": 2.365649938583374, "learning_rate": 0.0002, "epoch": 7.1894788114953725, "step": 7380}, {"loss": 0.6572, "grad_norm": 1.8271889686584473, "learning_rate": 0.0002, "epoch": 7.199220652703361, "step": 7390}, {"loss": 0.6955, "grad_norm": 1.9143747091293335, "learning_rate": 0.0002, "epoch": 7.208962493911349, "step": 7400}, {"loss": 0.6811, "grad_norm": 1.5670185089111328, "learning_rate": 0.0002, "epoch": 7.218704335119337, "step": 7410}, {"loss": 0.7035, "grad_norm": 1.7452768087387085, "learning_rate": 0.0002, "epoch": 7.228446176327326, "step": 7420}, {"loss": 0.6848, "grad_norm": 1.7830921411514282, "learning_rate": 0.0002, "epoch": 7.238188017535315, "step": 7430}, {"loss": 0.7018, "grad_norm": 1.9281501770019531, "learning_rate": 0.0002, "epoch": 7.247929858743302, "step": 7440}, {"loss": 0.6802, "grad_norm": 1.889663815498352, "learning_rate": 0.0002, "epoch": 7.257671699951291, "step": 7450}, {"loss": 0.6457, "grad_norm": 1.704999566078186, "learning_rate": 0.0002, "epoch": 7.2674135411592795, "step": 7460}, {"loss": 0.6919, "grad_norm": 1.824109435081482, "learning_rate": 0.0002, "epoch": 7.277155382367267, "step": 7470}, {"loss": 0.6814, "grad_norm": 1.5378915071487427, "learning_rate": 0.0002, "epoch": 7.286897223575256, "step": 7480}, {"loss": 0.7257, "grad_norm": 1.830587387084961, "learning_rate": 0.0002, "epoch": 7.296639064783244, "step": 7490}, {"loss": 0.7036, "grad_norm": 2.0029330253601074, "learning_rate": 0.0002, "epoch": 7.306380905991232, "step": 7500}, {"loss": 0.662, "grad_norm": 2.0871448516845703, "learning_rate": 0.0002, "epoch": 7.316122747199221, "step": 7510}, {"loss": 0.6856, "grad_norm": 1.8416074514389038, "learning_rate": 0.0002, "epoch": 7.325864588407209, "step": 7520}, {"loss": 0.688, "grad_norm": 1.8962771892547607, "learning_rate": 0.0002, "epoch": 7.335606429615197, "step": 7530}, {"loss": 0.6935, "grad_norm": 1.899487018585205, "learning_rate": 0.0002, "epoch": 7.345348270823186, "step": 7540}, {"loss": 0.667, "grad_norm": 1.8300765752792358, "learning_rate": 0.0002, "epoch": 7.355090112031174, "step": 7550}, {"loss": 0.7012, "grad_norm": 2.178112268447876, "learning_rate": 0.0002, "epoch": 7.364831953239162, "step": 7560}, {"loss": 0.6172, "grad_norm": 1.8472180366516113, "learning_rate": 0.0002, "epoch": 7.3745737944471506, "step": 7570}, {"loss": 0.6944, "grad_norm": 1.7787587642669678, "learning_rate": 0.0002, "epoch": 7.384315635655139, "step": 7580}, {"loss": 0.7236, "grad_norm": 1.8309564590454102, "learning_rate": 0.0002, "epoch": 7.394057476863127, "step": 7590}, {"loss": 0.7067, "grad_norm": 2.028923273086548, "learning_rate": 0.0002, "epoch": 7.4037993180711155, "step": 7600}, {"loss": 0.6656, "grad_norm": 1.7393525838851929, "learning_rate": 0.0002, "epoch": 7.413541159279104, "step": 7610}, {"loss": 0.6958, "grad_norm": 1.8816498517990112, "learning_rate": 0.0002, "epoch": 7.423283000487092, "step": 7620}, {"loss": 0.6666, "grad_norm": 2.4553585052490234, "learning_rate": 0.0002, "epoch": 7.43302484169508, "step": 7630}, {"loss": 0.6964, "grad_norm": 1.9045933485031128, "learning_rate": 0.0002, "epoch": 7.442766682903069, "step": 7640}, {"loss": 0.6759, "grad_norm": 1.664156198501587, "learning_rate": 0.0002, "epoch": 7.452508524111057, "step": 7650}, {"loss": 0.7777, "grad_norm": 1.792748212814331, "learning_rate": 0.0002, "epoch": 7.462250365319045, "step": 7660}, {"loss": 0.6785, "grad_norm": 1.8481247425079346, "learning_rate": 0.0002, "epoch": 7.471992206527034, "step": 7670}, {"loss": 0.7156, "grad_norm": 2.0541393756866455, "learning_rate": 0.0002, "epoch": 7.481734047735022, "step": 7680}, {"loss": 0.7126, "grad_norm": 1.594969630241394, "learning_rate": 0.0002, "epoch": 7.49147588894301, "step": 7690}, {"loss": 0.7042, "grad_norm": 2.1409924030303955, "learning_rate": 0.0002, "epoch": 7.501217730150999, "step": 7700}, {"loss": 0.6691, "grad_norm": 1.9743319749832153, "learning_rate": 0.0002, "epoch": 7.5109595713589865, "step": 7710}, {"loss": 0.677, "grad_norm": 1.866410493850708, "learning_rate": 0.0002, "epoch": 7.520701412566975, "step": 7720}, {"loss": 0.7159, "grad_norm": 1.9087774753570557, "learning_rate": 0.0002, "epoch": 7.530443253774964, "step": 7730}, {"loss": 0.7564, "grad_norm": 1.8624005317687988, "learning_rate": 0.0002, "epoch": 7.540185094982951, "step": 7740}, {"loss": 0.7355, "grad_norm": 1.629889726638794, "learning_rate": 0.0002, "epoch": 7.54992693619094, "step": 7750}, {"loss": 0.8001, "grad_norm": 2.1364638805389404, "learning_rate": 0.0002, "epoch": 7.559668777398929, "step": 7760}, {"loss": 0.7878, "grad_norm": 1.591701865196228, "learning_rate": 0.0002, "epoch": 7.569410618606916, "step": 7770}, {"loss": 0.7484, "grad_norm": 2.3200602531433105, "learning_rate": 0.0002, "epoch": 7.579152459814905, "step": 7780}, {"loss": 0.7407, "grad_norm": 1.9998793601989746, "learning_rate": 0.0002, "epoch": 7.5888943010228935, "step": 7790}, {"loss": 0.7436, "grad_norm": 1.8921900987625122, "learning_rate": 0.0002, "epoch": 7.598636142230882, "step": 7800}, {"loss": 0.6898, "grad_norm": 1.8826839923858643, "learning_rate": 0.0002, "epoch": 7.60837798343887, "step": 7810}, {"loss": 0.7376, "grad_norm": 1.8796452283859253, "learning_rate": 0.0002, "epoch": 7.618119824646858, "step": 7820}, {"loss": 0.7304, "grad_norm": 1.6528139114379883, "learning_rate": 0.0002, "epoch": 7.627861665854846, "step": 7830}, {"loss": 0.7205, "grad_norm": 1.9646536111831665, "learning_rate": 0.0002, "epoch": 7.637603507062835, "step": 7840}, {"loss": 0.7276, "grad_norm": 1.6951191425323486, "learning_rate": 0.0002, "epoch": 7.647345348270823, "step": 7850}, {"loss": 0.7476, "grad_norm": 1.8734302520751953, "learning_rate": 0.0002, "epoch": 7.657087189478812, "step": 7860}, {"loss": 0.729, "grad_norm": 2.140984058380127, "learning_rate": 0.0002, "epoch": 7.6668290306868, "step": 7870}, {"loss": 0.7979, "grad_norm": 1.8852670192718506, "learning_rate": 0.0002, "epoch": 7.676570871894788, "step": 7880}, {"loss": 0.745, "grad_norm": 2.1172003746032715, "learning_rate": 0.0002, "epoch": 7.686312713102776, "step": 7890}, {"loss": 0.7796, "grad_norm": 1.8237593173980713, "learning_rate": 0.0002, "epoch": 7.696054554310765, "step": 7900}, {"loss": 0.7203, "grad_norm": 2.1399245262145996, "learning_rate": 0.0002, "epoch": 7.705796395518753, "step": 7910}, {"loss": 0.741, "grad_norm": 1.8119547367095947, "learning_rate": 0.0002, "epoch": 7.715538236726742, "step": 7920}, {"loss": 0.7826, "grad_norm": 1.943442463874817, "learning_rate": 0.0002, "epoch": 7.7252800779347295, "step": 7930}, {"loss": 0.7635, "grad_norm": 1.6926734447479248, "learning_rate": 0.0002, "epoch": 7.735021919142718, "step": 7940}, {"loss": 0.7531, "grad_norm": 1.6824363470077515, "learning_rate": 0.0002, "epoch": 7.744763760350706, "step": 7950}, {"loss": 0.7203, "grad_norm": 1.8615055084228516, "learning_rate": 0.0002, "epoch": 7.754505601558694, "step": 7960}, {"loss": 0.7765, "grad_norm": 1.7171595096588135, "learning_rate": 0.0002, "epoch": 7.764247442766683, "step": 7970}, {"loss": 0.7633, "grad_norm": 1.9871152639389038, "learning_rate": 0.0002, "epoch": 7.773989283974672, "step": 7980}, {"loss": 0.8113, "grad_norm": 1.8975892066955566, "learning_rate": 0.0002, "epoch": 7.783731125182659, "step": 7990}, {"loss": 0.7363, "grad_norm": 1.8259385824203491, "learning_rate": 0.0002, "epoch": 7.793472966390648, "step": 8000}, {"loss": 0.7867, "grad_norm": 2.2361183166503906, "learning_rate": 0.0002, "epoch": 7.8032148075986365, "step": 8010}, {"loss": 0.7976, "grad_norm": 1.64067804813385, "learning_rate": 0.0002, "epoch": 7.812956648806624, "step": 8020}, {"loss": 0.7489, "grad_norm": 2.0037248134613037, "learning_rate": 0.0002, "epoch": 7.822698490014613, "step": 8030}, {"loss": 0.7219, "grad_norm": 1.8022961616516113, "learning_rate": 0.0002, "epoch": 7.832440331222601, "step": 8040}, {"loss": 0.7891, "grad_norm": 1.9980754852294922, "learning_rate": 0.0002, "epoch": 7.842182172430589, "step": 8050}, {"loss": 0.735, "grad_norm": 1.632716178894043, "learning_rate": 0.0002, "epoch": 7.851924013638578, "step": 8060}, {"loss": 0.7897, "grad_norm": 1.6348111629486084, "learning_rate": 0.0002, "epoch": 7.861665854846566, "step": 8070}, {"loss": 0.7585, "grad_norm": 1.968295693397522, "learning_rate": 0.0002, "epoch": 7.871407696054554, "step": 8080}, {"loss": 0.7531, "grad_norm": 1.6947685480117798, "learning_rate": 0.0002, "epoch": 7.881149537262543, "step": 8090}, {"loss": 0.7828, "grad_norm": 6.1600341796875, "learning_rate": 0.0002, "epoch": 7.890891378470531, "step": 8100}, {"loss": 0.7612, "grad_norm": 1.9334033727645874, "learning_rate": 0.0002, "epoch": 7.900633219678519, "step": 8110}, {"loss": 0.7877, "grad_norm": 1.729058027267456, "learning_rate": 0.0002, "epoch": 7.9103750608865075, "step": 8120}, {"loss": 0.788, "grad_norm": 1.8671422004699707, "learning_rate": 0.0002, "epoch": 7.920116902094496, "step": 8130}, {"loss": 0.7798, "grad_norm": 1.9794875383377075, "learning_rate": 0.0002, "epoch": 7.929858743302484, "step": 8140}, {"loss": 0.7705, "grad_norm": 1.812229037284851, "learning_rate": 0.0002, "epoch": 7.9396005845104725, "step": 8150}, {"loss": 0.8335, "grad_norm": 1.7354048490524292, "learning_rate": 0.0002, "epoch": 7.949342425718461, "step": 8160}, {"loss": 0.7849, "grad_norm": 1.7386713027954102, "learning_rate": 0.0002, "epoch": 7.959084266926449, "step": 8170}, {"loss": 0.7829, "grad_norm": 1.917111873626709, "learning_rate": 0.0002, "epoch": 7.968826108134437, "step": 8180}, {"loss": 0.7672, "grad_norm": 1.7007793188095093, "learning_rate": 0.0002, "epoch": 7.978567949342426, "step": 8190}, {"loss": 0.8181, "grad_norm": 1.8241386413574219, "learning_rate": 0.0002, "epoch": 7.988309790550414, "step": 8200}]}