diff --git a/.gitattributes b/.gitattributes
index 8f8dd6d145f5c1752bb92b8431115cf5432440b1..5004930cb5655e4ae59473d80bd07245d91f85b6 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -866,3 +866,12 @@ gemma-2b-it_int4_winogrande-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-
 gemma-2b-it_int4_winogrande-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-798-sd-4/checkpoint-49/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_winogrande-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-798-sd-4/checkpoint-98/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_winogrande-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-798-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9a25de9298ae67ee0c6a132a503d40bed8ca8c54
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0e46139fbdbe7c9dba8da0dc7642049d7fa284650682279a059b4fd5f0e13ba
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7bd77e87dedf76c787f4902bc2dd2c5d416b5bfe
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82cb608022eb1e5aa4f5ec3ee9388bb9bea9fb6045e2196e29dd6bbb62b0a9e7
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..524eb41dbb8f8fe2db2b0f0bedb5a885ace85621
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b30105b3780bd608eb8fdffd04d7d5b7c9e01bb1f83899e5684f65b26608103f
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3ca5d1b71495d675490231c16589dcd481f0db2e
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40504680aa3f8504409b92ec31449b9e1b0a70e5047c0cc85fdc9743e92922cc
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6a00200e9ee54972a74baebcb9bb5376cec0bb86
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7e3656ae7e8e59bbccc6990a75d6003a70342a36e1351fe42c68581b546391f
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..824ce29fe6338fd8c98d60591ec6be99311b1796
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/trainer_state.json
@@ -0,0 +1,7586 @@
+{
+  "best_metric": 1.4113320112228394,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120",
+  "epoch": 7.0,
+  "eval_steps": 10,
+  "global_step": 10710,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0065359477124183,
+      "grad_norm": 0.6926420331001282,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0130718954248366,
+      "grad_norm": 0.6656355857849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 4610
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 0.9901936650276184,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 4620
+    },
+    {
+      "epoch": 3.026143790849673,
+      "grad_norm": 0.6713474988937378,
+      "learning_rate": 0.0002,
+      "loss": 1.22,
+      "step": 4630
+    },
+    {
+      "epoch": 3.0326797385620914,
+      "grad_norm": 0.6199324131011963,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 4640
+    },
+    {
+      "epoch": 3.0392156862745097,
+      "grad_norm": 0.7180785536766052,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 4650
+    },
+    {
+      "epoch": 3.045751633986928,
+      "grad_norm": 0.8256588578224182,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 4660
+    },
+    {
+      "epoch": 3.052287581699346,
+      "grad_norm": 0.6637389063835144,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 4670
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 0.6980698108673096,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 4680
+    },
+    {
+      "epoch": 3.065359477124183,
+      "grad_norm": 0.8091534972190857,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 4690
+    },
+    {
+      "epoch": 3.0718954248366015,
+      "grad_norm": 0.5715174078941345,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 4700
+    },
+    {
+      "epoch": 3.0784313725490198,
+      "grad_norm": 0.735639750957489,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 4710
+    },
+    {
+      "epoch": 3.084967320261438,
+      "grad_norm": 0.7619708180427551,
+      "learning_rate": 0.0002,
+      "loss": 1.1522,
+      "step": 4720
+    },
+    {
+      "epoch": 3.0915032679738563,
+      "grad_norm": 1.263566017150879,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 4730
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 0.6600871682167053,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4740
+    },
+    {
+      "epoch": 3.104575163398693,
+      "grad_norm": 0.717792809009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 4750
+    },
+    {
+      "epoch": 3.111111111111111,
+      "grad_norm": 0.853714644908905,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1176470588235294,
+      "grad_norm": 1.1004153490066528,
+      "learning_rate": 0.0002,
+      "loss": 1.2031,
+      "step": 4770
+    },
+    {
+      "epoch": 3.1241830065359477,
+      "grad_norm": 0.8566235899925232,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 4780
+    },
+    {
+      "epoch": 3.130718954248366,
+      "grad_norm": 0.8315296173095703,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 4790
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 0.8020524978637695,
+      "learning_rate": 0.0002,
+      "loss": 1.32,
+      "step": 4800
+    },
+    {
+      "epoch": 3.1437908496732025,
+      "grad_norm": 0.7564275860786438,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 4810
+    },
+    {
+      "epoch": 3.150326797385621,
+      "grad_norm": 0.9077776670455933,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 4820
+    },
+    {
+      "epoch": 3.156862745098039,
+      "grad_norm": 0.6323099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 4830
+    },
+    {
+      "epoch": 3.1633986928104574,
+      "grad_norm": 0.6625368595123291,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 4840
+    },
+    {
+      "epoch": 3.1699346405228757,
+      "grad_norm": 0.8119261860847473,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 4850
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 0.6399450898170471,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 4860
+    },
+    {
+      "epoch": 3.183006535947712,
+      "grad_norm": 1.0659016370773315,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 4870
+    },
+    {
+      "epoch": 3.189542483660131,
+      "grad_norm": 0.8040369749069214,
+      "learning_rate": 0.0002,
+      "loss": 1.2914,
+      "step": 4880
+    },
+    {
+      "epoch": 3.196078431372549,
+      "grad_norm": 0.7784733176231384,
+      "learning_rate": 0.0002,
+      "loss": 1.1996,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2026143790849675,
+      "grad_norm": 0.9660294651985168,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 4900
+    },
+    {
+      "epoch": 3.2091503267973858,
+      "grad_norm": 1.0676977634429932,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 4910
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 0.5877565741539001,
+      "learning_rate": 0.0002,
+      "loss": 1.0083,
+      "step": 4920
+    },
+    {
+      "epoch": 3.2222222222222223,
+      "grad_norm": 0.6164032816886902,
+      "learning_rate": 0.0002,
+      "loss": 1.1046,
+      "step": 4930
+    },
+    {
+      "epoch": 3.2287581699346406,
+      "grad_norm": 0.7627606987953186,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 4940
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.7442803978919983,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 4950
+    },
+    {
+      "epoch": 3.241830065359477,
+      "grad_norm": 0.7277812361717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 4960
+    },
+    {
+      "epoch": 3.2483660130718954,
+      "grad_norm": 1.0301902294158936,
+      "learning_rate": 0.0002,
+      "loss": 1.2237,
+      "step": 4970
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 0.7798232436180115,
+      "learning_rate": 0.0002,
+      "loss": 1.1466,
+      "step": 4980
+    },
+    {
+      "epoch": 3.261437908496732,
+      "grad_norm": 1.210265874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.2142,
+      "step": 4990
+    },
+    {
+      "epoch": 3.2679738562091503,
+      "grad_norm": 0.6677713990211487,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2745098039215685,
+      "grad_norm": 1.0524500608444214,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 5010
+    },
+    {
+      "epoch": 3.281045751633987,
+      "grad_norm": 0.7091745734214783,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5020
+    },
+    {
+      "epoch": 3.287581699346405,
+      "grad_norm": 0.8523224592208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1891,
+      "step": 5030
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 0.6120608448982239,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3006535947712417,
+      "grad_norm": 0.7437472939491272,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 5050
+    },
+    {
+      "epoch": 3.30718954248366,
+      "grad_norm": 0.7611715197563171,
+      "learning_rate": 0.0002,
+      "loss": 1.1295,
+      "step": 5060
+    },
+    {
+      "epoch": 3.313725490196078,
+      "grad_norm": 0.7249704003334045,
+      "learning_rate": 0.0002,
+      "loss": 1.0531,
+      "step": 5070
+    },
+    {
+      "epoch": 3.3202614379084965,
+      "grad_norm": 0.7316247820854187,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5080
+    },
+    {
+      "epoch": 3.326797385620915,
+      "grad_norm": 0.562412440776825,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5090
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.7052176594734192,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5100
+    },
+    {
+      "epoch": 3.3398692810457518,
+      "grad_norm": 0.7714211344718933,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 5110
+    },
+    {
+      "epoch": 3.34640522875817,
+      "grad_norm": 1.0436055660247803,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5120
+    },
+    {
+      "epoch": 3.3529411764705883,
+      "grad_norm": 0.8867271542549133,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 5130
+    },
+    {
+      "epoch": 3.3594771241830066,
+      "grad_norm": 0.8371267914772034,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 5140
+    },
+    {
+      "epoch": 3.366013071895425,
+      "grad_norm": 0.7257837057113647,
+      "learning_rate": 0.0002,
+      "loss": 1.1073,
+      "step": 5150
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 0.7102002501487732,
+      "learning_rate": 0.0002,
+      "loss": 1.1162,
+      "step": 5160
+    },
+    {
+      "epoch": 3.3790849673202614,
+      "grad_norm": 0.7636350393295288,
+      "learning_rate": 0.0002,
+      "loss": 1.2056,
+      "step": 5170
+    },
+    {
+      "epoch": 3.3856209150326797,
+      "grad_norm": 0.6887359619140625,
+      "learning_rate": 0.0002,
+      "loss": 1.0708,
+      "step": 5180
+    },
+    {
+      "epoch": 3.392156862745098,
+      "grad_norm": 0.8141424655914307,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 5190
+    },
+    {
+      "epoch": 3.3986928104575163,
+      "grad_norm": 0.694423496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5200
+    },
+    {
+      "epoch": 3.4052287581699345,
+      "grad_norm": 0.914013683795929,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5210
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 0.8503239750862122,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 5220
+    },
+    {
+      "epoch": 3.418300653594771,
+      "grad_norm": 0.6196836233139038,
+      "learning_rate": 0.0002,
+      "loss": 1.2696,
+      "step": 5230
+    },
+    {
+      "epoch": 3.4248366013071894,
+      "grad_norm": 1.0760811567306519,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 5240
+    },
+    {
+      "epoch": 3.431372549019608,
+      "grad_norm": 0.6524698138237,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 5250
+    },
+    {
+      "epoch": 3.4379084967320264,
+      "grad_norm": 0.674467921257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5260
+    },
+    {
+      "epoch": 3.4444444444444446,
+      "grad_norm": 0.7690372467041016,
+      "learning_rate": 0.0002,
+      "loss": 1.1015,
+      "step": 5270
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 0.8751813769340515,
+      "learning_rate": 0.0002,
+      "loss": 1.2511,
+      "step": 5280
+    },
+    {
+      "epoch": 3.457516339869281,
+      "grad_norm": 0.750407874584198,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 5290
+    },
+    {
+      "epoch": 3.4640522875816995,
+      "grad_norm": 0.5991823077201843,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 5300
+    },
+    {
+      "epoch": 3.4705882352941178,
+      "grad_norm": 1.0164772272109985,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 5310
+    },
+    {
+      "epoch": 3.477124183006536,
+      "grad_norm": 0.8704105019569397,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 5320
+    },
+    {
+      "epoch": 3.4836601307189543,
+      "grad_norm": 0.709102213382721,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 5330
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 0.6273632049560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 5340
+    },
+    {
+      "epoch": 3.496732026143791,
+      "grad_norm": 0.6807359457015991,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 5350
+    },
+    {
+      "epoch": 3.503267973856209,
+      "grad_norm": 0.7085188627243042,
+      "learning_rate": 0.0002,
+      "loss": 1.131,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5098039215686274,
+      "grad_norm": 0.6938307881355286,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 5370
+    },
+    {
+      "epoch": 3.5163398692810457,
+      "grad_norm": 0.8544146418571472,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 5380
+    },
+    {
+      "epoch": 3.522875816993464,
+      "grad_norm": 0.7889642119407654,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 5390
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.7858421206474304,
+      "learning_rate": 0.0002,
+      "loss": 1.1691,
+      "step": 5400
+    },
+    {
+      "epoch": 3.5359477124183005,
+      "grad_norm": 0.8547123074531555,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5410
+    },
+    {
+      "epoch": 3.542483660130719,
+      "grad_norm": 0.8218181133270264,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 5420
+    },
+    {
+      "epoch": 3.549019607843137,
+      "grad_norm": 1.153623342514038,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 5430
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.1321099996566772,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 5440
+    },
+    {
+      "epoch": 3.5620915032679736,
+      "grad_norm": 0.9495334029197693,
+      "learning_rate": 0.0002,
+      "loss": 1.2224,
+      "step": 5450
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 0.8743821978569031,
+      "learning_rate": 0.0002,
+      "loss": 1.2869,
+      "step": 5460
+    },
+    {
+      "epoch": 3.57516339869281,
+      "grad_norm": 0.7513086795806885,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 5470
+    },
+    {
+      "epoch": 3.581699346405229,
+      "grad_norm": 1.0139480829238892,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 5480
+    },
+    {
+      "epoch": 3.588235294117647,
+      "grad_norm": 0.6615135073661804,
+      "learning_rate": 0.0002,
+      "loss": 1.1706,
+      "step": 5490
+    },
+    {
+      "epoch": 3.5947712418300655,
+      "grad_norm": 1.180798888206482,
+      "learning_rate": 0.0002,
+      "loss": 1.3906,
+      "step": 5500
+    },
+    {
+      "epoch": 3.6013071895424837,
+      "grad_norm": 0.7085279226303101,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 5510
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 0.540268063545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1623,
+      "step": 5520
+    },
+    {
+      "epoch": 3.6143790849673203,
+      "grad_norm": 0.7905671000480652,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 5530
+    },
+    {
+      "epoch": 3.6209150326797386,
+      "grad_norm": 0.8457717299461365,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 5540
+    },
+    {
+      "epoch": 3.627450980392157,
+      "grad_norm": 0.7102677822113037,
+      "learning_rate": 0.0002,
+      "loss": 1.1799,
+      "step": 5550
+    },
+    {
+      "epoch": 3.633986928104575,
+      "grad_norm": 0.7179514765739441,
+      "learning_rate": 0.0002,
+      "loss": 1.2394,
+      "step": 5560
+    },
+    {
+      "epoch": 3.6405228758169934,
+      "grad_norm": 1.0854148864746094,
+      "learning_rate": 0.0002,
+      "loss": 1.2019,
+      "step": 5570
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 0.8209951519966125,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5580
+    },
+    {
+      "epoch": 3.65359477124183,
+      "grad_norm": 0.6944138407707214,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 5590
+    },
+    {
+      "epoch": 3.6601307189542482,
+      "grad_norm": 0.7675473093986511,
+      "learning_rate": 0.0002,
+      "loss": 1.3226,
+      "step": 5600
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6683364510536194,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 5610
+    },
+    {
+      "epoch": 3.6732026143790852,
+      "grad_norm": 0.7920727133750916,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 5620
+    },
+    {
+      "epoch": 3.6797385620915035,
+      "grad_norm": 0.9440218806266785,
+      "learning_rate": 0.0002,
+      "loss": 1.2287,
+      "step": 5630
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 0.6600824594497681,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 5640
+    },
+    {
+      "epoch": 3.69281045751634,
+      "grad_norm": 0.6860619187355042,
+      "learning_rate": 0.0002,
+      "loss": 1.191,
+      "step": 5650
+    },
+    {
+      "epoch": 3.6993464052287583,
+      "grad_norm": 0.6579713225364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 5660
+    },
+    {
+      "epoch": 3.7058823529411766,
+      "grad_norm": 0.661081075668335,
+      "learning_rate": 0.0002,
+      "loss": 1.1464,
+      "step": 5670
+    },
+    {
+      "epoch": 3.712418300653595,
+      "grad_norm": 1.0968825817108154,
+      "learning_rate": 0.0002,
+      "loss": 1.289,
+      "step": 5680
+    },
+    {
+      "epoch": 3.718954248366013,
+      "grad_norm": 0.8066844940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.192,
+      "step": 5690
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 0.8341682553291321,
+      "learning_rate": 0.0002,
+      "loss": 1.2322,
+      "step": 5700
+    },
+    {
+      "epoch": 3.7320261437908497,
+      "grad_norm": 0.6682852506637573,
+      "learning_rate": 0.0002,
+      "loss": 1.1473,
+      "step": 5710
+    },
+    {
+      "epoch": 3.738562091503268,
+      "grad_norm": 0.898595929145813,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 5720
+    },
+    {
+      "epoch": 3.7450980392156863,
+      "grad_norm": 0.6876054406166077,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 5730
+    },
+    {
+      "epoch": 3.7516339869281046,
+      "grad_norm": 0.7817103266716003,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 5740
+    },
+    {
+      "epoch": 3.758169934640523,
+      "grad_norm": 0.5840168595314026,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 5750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 0.6263918876647949,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5760
+    },
+    {
+      "epoch": 3.7712418300653594,
+      "grad_norm": 0.7948952317237854,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 5770
+    },
+    {
+      "epoch": 3.7777777777777777,
+      "grad_norm": 0.6700998544692993,
+      "learning_rate": 0.0002,
+      "loss": 1.149,
+      "step": 5780
+    },
+    {
+      "epoch": 3.784313725490196,
+      "grad_norm": 1.1169519424438477,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 5790
+    },
+    {
+      "epoch": 3.7908496732026142,
+      "grad_norm": 0.8354471325874329,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 5800
+    },
+    {
+      "epoch": 3.7973856209150325,
+      "grad_norm": 0.6304181814193726,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 5810
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 0.6919655799865723,
+      "learning_rate": 0.0002,
+      "loss": 1.2059,
+      "step": 5820
+    },
+    {
+      "epoch": 3.810457516339869,
+      "grad_norm": 0.600385844707489,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5830
+    },
+    {
+      "epoch": 3.8169934640522873,
+      "grad_norm": 0.8406319618225098,
+      "learning_rate": 0.0002,
+      "loss": 1.2324,
+      "step": 5840
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.7594282031059265,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5850
+    },
+    {
+      "epoch": 3.8300653594771243,
+      "grad_norm": 0.8179879784584045,
+      "learning_rate": 0.0002,
+      "loss": 1.1903,
+      "step": 5860
+    },
+    {
+      "epoch": 3.8366013071895426,
+      "grad_norm": 1.141430377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 5870
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 0.6595550775527954,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 5880
+    },
+    {
+      "epoch": 3.849673202614379,
+      "grad_norm": 0.7499435544013977,
+      "learning_rate": 0.0002,
+      "loss": 1.2378,
+      "step": 5890
+    },
+    {
+      "epoch": 3.8562091503267975,
+      "grad_norm": 0.7851517200469971,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5900
+    },
+    {
+      "epoch": 3.8627450980392157,
+      "grad_norm": 1.0533545017242432,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 5910
+    },
+    {
+      "epoch": 3.869281045751634,
+      "grad_norm": 0.960086464881897,
+      "learning_rate": 0.0002,
+      "loss": 1.3576,
+      "step": 5920
+    },
+    {
+      "epoch": 3.8758169934640523,
+      "grad_norm": 0.9952049851417542,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 5930
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 0.7884191274642944,
+      "learning_rate": 0.0002,
+      "loss": 1.2027,
+      "step": 5940
+    },
+    {
+      "epoch": 3.888888888888889,
+      "grad_norm": 0.7461766600608826,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5950
+    },
+    {
+      "epoch": 3.895424836601307,
+      "grad_norm": 0.9594355821609497,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 5960
+    },
+    {
+      "epoch": 3.9019607843137254,
+      "grad_norm": 0.8179471492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.1164,
+      "step": 5970
+    },
+    {
+      "epoch": 3.9084967320261437,
+      "grad_norm": 0.8240267634391785,
+      "learning_rate": 0.0002,
+      "loss": 1.2421,
+      "step": 5980
+    },
+    {
+      "epoch": 3.915032679738562,
+      "grad_norm": 0.7462618350982666,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 5990
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 0.711207389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.2124,
+      "step": 6000
+    },
+    {
+      "epoch": 3.928104575163399,
+      "grad_norm": 0.6910956501960754,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 6010
+    },
+    {
+      "epoch": 3.9346405228758172,
+      "grad_norm": 0.749093770980835,
+      "learning_rate": 0.0002,
+      "loss": 1.2127,
+      "step": 6020
+    },
+    {
+      "epoch": 3.9411764705882355,
+      "grad_norm": 1.3332762718200684,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6030
+    },
+    {
+      "epoch": 3.947712418300654,
+      "grad_norm": 0.71457439661026,
+      "learning_rate": 0.0002,
+      "loss": 1.1442,
+      "step": 6040
+    },
+    {
+      "epoch": 3.954248366013072,
+      "grad_norm": 1.1205238103866577,
+      "learning_rate": 0.0002,
+      "loss": 1.339,
+      "step": 6050
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 0.6958928108215332,
+      "learning_rate": 0.0002,
+      "loss": 1.2962,
+      "step": 6060
+    },
+    {
+      "epoch": 3.9673202614379086,
+      "grad_norm": 0.7518056035041809,
+      "learning_rate": 0.0002,
+      "loss": 1.1802,
+      "step": 6070
+    },
+    {
+      "epoch": 3.973856209150327,
+      "grad_norm": 0.8010755777359009,
+      "learning_rate": 0.0002,
+      "loss": 1.1179,
+      "step": 6080
+    },
+    {
+      "epoch": 3.980392156862745,
+      "grad_norm": 0.7492658495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 6090
+    },
+    {
+      "epoch": 3.9869281045751634,
+      "grad_norm": 0.900704562664032,
+      "learning_rate": 0.0002,
+      "loss": 1.2113,
+      "step": 6100
+    },
+    {
+      "epoch": 3.9934640522875817,
+      "grad_norm": 0.7997331619262695,
+      "learning_rate": 0.0002,
+      "loss": 1.1106,
+      "step": 6110
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7163209319114685,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 6120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4113320112228394,
+      "eval_runtime": 33.7199,
+      "eval_samples_per_second": 12.93,
+      "eval_steps_per_second": 1.631,
+      "step": 6120
+    },
+    {
+      "epoch": 4.006535947712418,
+      "grad_norm": 0.9527022838592529,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 6130
+    },
+    {
+      "epoch": 4.0130718954248366,
+      "grad_norm": 0.7603210210800171,
+      "learning_rate": 0.0002,
+      "loss": 1.101,
+      "step": 6140
+    },
+    {
+      "epoch": 4.019607843137255,
+      "grad_norm": 1.127387523651123,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 6150
+    },
+    {
+      "epoch": 4.026143790849673,
+      "grad_norm": 0.8290133476257324,
+      "learning_rate": 0.0002,
+      "loss": 1.0734,
+      "step": 6160
+    },
+    {
+      "epoch": 4.032679738562091,
+      "grad_norm": 0.9912241101264954,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 6170
+    },
+    {
+      "epoch": 4.03921568627451,
+      "grad_norm": 0.947005033493042,
+      "learning_rate": 0.0002,
+      "loss": 1.0719,
+      "step": 6180
+    },
+    {
+      "epoch": 4.045751633986928,
+      "grad_norm": 0.707466185092926,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 6190
+    },
+    {
+      "epoch": 4.052287581699346,
+      "grad_norm": 1.0604327917099,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6200
+    },
+    {
+      "epoch": 4.0588235294117645,
+      "grad_norm": 0.7848685383796692,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6210
+    },
+    {
+      "epoch": 4.065359477124183,
+      "grad_norm": 0.8475256562232971,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 6220
+    },
+    {
+      "epoch": 4.071895424836601,
+      "grad_norm": 0.9759448766708374,
+      "learning_rate": 0.0002,
+      "loss": 1.1104,
+      "step": 6230
+    },
+    {
+      "epoch": 4.078431372549019,
+      "grad_norm": 0.9324519038200378,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 6240
+    },
+    {
+      "epoch": 4.084967320261438,
+      "grad_norm": 0.8723901510238647,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 6250
+    },
+    {
+      "epoch": 4.091503267973856,
+      "grad_norm": 0.8343415856361389,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6260
+    },
+    {
+      "epoch": 4.098039215686274,
+      "grad_norm": 0.7490310072898865,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 6270
+    },
+    {
+      "epoch": 4.104575163398692,
+      "grad_norm": 0.8961182832717896,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 6280
+    },
+    {
+      "epoch": 4.111111111111111,
+      "grad_norm": 0.7124854922294617,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 6290
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 0.8338138461112976,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 6300
+    },
+    {
+      "epoch": 4.124183006535947,
+      "grad_norm": 0.8075833320617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 6310
+    },
+    {
+      "epoch": 4.130718954248366,
+      "grad_norm": 0.8069391846656799,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 6320
+    },
+    {
+      "epoch": 4.137254901960785,
+      "grad_norm": 0.9567893147468567,
+      "learning_rate": 0.0002,
+      "loss": 0.948,
+      "step": 6330
+    },
+    {
+      "epoch": 4.143790849673203,
+      "grad_norm": 1.2184662818908691,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 6340
+    },
+    {
+      "epoch": 4.150326797385621,
+      "grad_norm": 1.030976414680481,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 6350
+    },
+    {
+      "epoch": 4.1568627450980395,
+      "grad_norm": 0.9749957323074341,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6360
+    },
+    {
+      "epoch": 4.163398692810458,
+      "grad_norm": 0.7089483141899109,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 6370
+    },
+    {
+      "epoch": 4.169934640522876,
+      "grad_norm": 1.1084946393966675,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 6380
+    },
+    {
+      "epoch": 4.176470588235294,
+      "grad_norm": 0.7998497486114502,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 6390
+    },
+    {
+      "epoch": 4.183006535947713,
+      "grad_norm": 0.8997811675071716,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 6400
+    },
+    {
+      "epoch": 4.189542483660131,
+      "grad_norm": 0.8359479904174805,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 6410
+    },
+    {
+      "epoch": 4.196078431372549,
+      "grad_norm": 0.9087472558021545,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 6420
+    },
+    {
+      "epoch": 4.2026143790849675,
+      "grad_norm": 1.1100451946258545,
+      "learning_rate": 0.0002,
+      "loss": 1.0657,
+      "step": 6430
+    },
+    {
+      "epoch": 4.209150326797386,
+      "grad_norm": 0.9376999735832214,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 6440
+    },
+    {
+      "epoch": 4.215686274509804,
+      "grad_norm": 0.8179266452789307,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 6450
+    },
+    {
+      "epoch": 4.222222222222222,
+      "grad_norm": 0.9953271746635437,
+      "learning_rate": 0.0002,
+      "loss": 1.0679,
+      "step": 6460
+    },
+    {
+      "epoch": 4.228758169934641,
+      "grad_norm": 0.8476650714874268,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 6470
+    },
+    {
+      "epoch": 4.235294117647059,
+      "grad_norm": 0.8406323194503784,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 6480
+    },
+    {
+      "epoch": 4.241830065359477,
+      "grad_norm": 0.819134533405304,
+      "learning_rate": 0.0002,
+      "loss": 1.057,
+      "step": 6490
+    },
+    {
+      "epoch": 4.248366013071895,
+      "grad_norm": 0.7764983773231506,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 6500
+    },
+    {
+      "epoch": 4.254901960784314,
+      "grad_norm": 0.8252112865447998,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 6510
+    },
+    {
+      "epoch": 4.261437908496732,
+      "grad_norm": 0.7941019535064697,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 6520
+    },
+    {
+      "epoch": 4.26797385620915,
+      "grad_norm": 0.7673905491828918,
+      "learning_rate": 0.0002,
+      "loss": 1.0296,
+      "step": 6530
+    },
+    {
+      "epoch": 4.2745098039215685,
+      "grad_norm": 0.8749890327453613,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6540
+    },
+    {
+      "epoch": 4.281045751633987,
+      "grad_norm": 0.7343207597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 6550
+    },
+    {
+      "epoch": 4.287581699346405,
+      "grad_norm": 1.2786651849746704,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6560
+    },
+    {
+      "epoch": 4.294117647058823,
+      "grad_norm": 1.316875696182251,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6570
+    },
+    {
+      "epoch": 4.300653594771242,
+      "grad_norm": 0.8349189162254333,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 6580
+    },
+    {
+      "epoch": 4.30718954248366,
+      "grad_norm": 0.7510647177696228,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6590
+    },
+    {
+      "epoch": 4.313725490196078,
+      "grad_norm": 0.932420551776886,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6600
+    },
+    {
+      "epoch": 4.3202614379084965,
+      "grad_norm": 0.8510616421699524,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 6610
+    },
+    {
+      "epoch": 4.326797385620915,
+      "grad_norm": 0.7661547064781189,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6620
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 1.0370930433273315,
+      "learning_rate": 0.0002,
+      "loss": 1.2064,
+      "step": 6630
+    },
+    {
+      "epoch": 4.339869281045751,
+      "grad_norm": 0.9302158951759338,
+      "learning_rate": 0.0002,
+      "loss": 1.1064,
+      "step": 6640
+    },
+    {
+      "epoch": 4.34640522875817,
+      "grad_norm": 0.9203811883926392,
+      "learning_rate": 0.0002,
+      "loss": 0.968,
+      "step": 6650
+    },
+    {
+      "epoch": 4.352941176470588,
+      "grad_norm": 0.9986332654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 6660
+    },
+    {
+      "epoch": 4.359477124183006,
+      "grad_norm": 0.8001713156700134,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6670
+    },
+    {
+      "epoch": 4.366013071895424,
+      "grad_norm": 0.829714298248291,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 6680
+    },
+    {
+      "epoch": 4.372549019607844,
+      "grad_norm": 0.8253079056739807,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 6690
+    },
+    {
+      "epoch": 4.379084967320262,
+      "grad_norm": 0.824666440486908,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 6700
+    },
+    {
+      "epoch": 4.38562091503268,
+      "grad_norm": 0.8872972130775452,
+      "learning_rate": 0.0002,
+      "loss": 1.1968,
+      "step": 6710
+    },
+    {
+      "epoch": 4.392156862745098,
+      "grad_norm": 0.8729761838912964,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 6720
+    },
+    {
+      "epoch": 4.398692810457517,
+      "grad_norm": 1.1367264986038208,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 6730
+    },
+    {
+      "epoch": 4.405228758169935,
+      "grad_norm": 0.9699058532714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0184,
+      "step": 6740
+    },
+    {
+      "epoch": 4.411764705882353,
+      "grad_norm": 0.8266763687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 6750
+    },
+    {
+      "epoch": 4.4183006535947715,
+      "grad_norm": 1.0249767303466797,
+      "learning_rate": 0.0002,
+      "loss": 1.0735,
+      "step": 6760
+    },
+    {
+      "epoch": 4.42483660130719,
+      "grad_norm": 0.73606938123703,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 6770
+    },
+    {
+      "epoch": 4.431372549019608,
+      "grad_norm": 1.4050679206848145,
+      "learning_rate": 0.0002,
+      "loss": 1.1037,
+      "step": 6780
+    },
+    {
+      "epoch": 4.437908496732026,
+      "grad_norm": 1.1114081144332886,
+      "learning_rate": 0.0002,
+      "loss": 1.1418,
+      "step": 6790
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8031067848205566,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 6800
+    },
+    {
+      "epoch": 4.450980392156863,
+      "grad_norm": 0.8513566851615906,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6810
+    },
+    {
+      "epoch": 4.457516339869281,
+      "grad_norm": 1.332741379737854,
+      "learning_rate": 0.0002,
+      "loss": 1.1852,
+      "step": 6820
+    },
+    {
+      "epoch": 4.4640522875816995,
+      "grad_norm": 1.5032578706741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 6830
+    },
+    {
+      "epoch": 4.470588235294118,
+      "grad_norm": 0.7677283883094788,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6840
+    },
+    {
+      "epoch": 4.477124183006536,
+      "grad_norm": 0.989148736000061,
+      "learning_rate": 0.0002,
+      "loss": 1.1501,
+      "step": 6850
+    },
+    {
+      "epoch": 4.483660130718954,
+      "grad_norm": 1.5316275358200073,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 6860
+    },
+    {
+      "epoch": 4.490196078431373,
+      "grad_norm": 0.9427124261856079,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 6870
+    },
+    {
+      "epoch": 4.496732026143791,
+      "grad_norm": 1.215287685394287,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 6880
+    },
+    {
+      "epoch": 4.503267973856209,
+      "grad_norm": 0.7286760210990906,
+      "learning_rate": 0.0002,
+      "loss": 1.0809,
+      "step": 6890
+    },
+    {
+      "epoch": 4.509803921568627,
+      "grad_norm": 0.874829888343811,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6900
+    },
+    {
+      "epoch": 4.516339869281046,
+      "grad_norm": 0.8058359622955322,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 6910
+    },
+    {
+      "epoch": 4.522875816993464,
+      "grad_norm": 1.248195767402649,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 6920
+    },
+    {
+      "epoch": 4.529411764705882,
+      "grad_norm": 0.8033645749092102,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 6930
+    },
+    {
+      "epoch": 4.5359477124183005,
+      "grad_norm": 1.7361950874328613,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 6940
+    },
+    {
+      "epoch": 4.542483660130719,
+      "grad_norm": 0.8058095574378967,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 6950
+    },
+    {
+      "epoch": 4.549019607843137,
+      "grad_norm": 1.254089593887329,
+      "learning_rate": 0.0002,
+      "loss": 1.0057,
+      "step": 6960
+    },
+    {
+      "epoch": 4.555555555555555,
+      "grad_norm": 0.9180455803871155,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 6970
+    },
+    {
+      "epoch": 4.562091503267974,
+      "grad_norm": 0.6677682399749756,
+      "learning_rate": 0.0002,
+      "loss": 1.0559,
+      "step": 6980
+    },
+    {
+      "epoch": 4.568627450980392,
+      "grad_norm": 0.8127354383468628,
+      "learning_rate": 0.0002,
+      "loss": 1.0453,
+      "step": 6990
+    },
+    {
+      "epoch": 4.57516339869281,
+      "grad_norm": 1.0263001918792725,
+      "learning_rate": 0.0002,
+      "loss": 1.0828,
+      "step": 7000
+    },
+    {
+      "epoch": 4.5816993464052285,
+      "grad_norm": 0.9641909003257751,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 7010
+    },
+    {
+      "epoch": 4.588235294117647,
+      "grad_norm": 0.9440861344337463,
+      "learning_rate": 0.0002,
+      "loss": 1.179,
+      "step": 7020
+    },
+    {
+      "epoch": 4.594771241830065,
+      "grad_norm": 0.9539011716842651,
+      "learning_rate": 0.0002,
+      "loss": 1.0931,
+      "step": 7030
+    },
+    {
+      "epoch": 4.601307189542483,
+      "grad_norm": 1.0449910163879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7040
+    },
+    {
+      "epoch": 4.607843137254902,
+      "grad_norm": 0.8766893744468689,
+      "learning_rate": 0.0002,
+      "loss": 0.9944,
+      "step": 7050
+    },
+    {
+      "epoch": 4.61437908496732,
+      "grad_norm": 0.6983462572097778,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 7060
+    },
+    {
+      "epoch": 4.620915032679738,
+      "grad_norm": 0.9505505561828613,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 7070
+    },
+    {
+      "epoch": 4.627450980392156,
+      "grad_norm": 1.2506657838821411,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7080
+    },
+    {
+      "epoch": 4.633986928104575,
+      "grad_norm": 0.9602801203727722,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 7090
+    },
+    {
+      "epoch": 4.640522875816993,
+      "grad_norm": 0.7398977875709534,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 7100
+    },
+    {
+      "epoch": 4.647058823529412,
+      "grad_norm": 1.3862425088882446,
+      "learning_rate": 0.0002,
+      "loss": 1.0769,
+      "step": 7110
+    },
+    {
+      "epoch": 4.65359477124183,
+      "grad_norm": 1.1451990604400635,
+      "learning_rate": 0.0002,
+      "loss": 1.0571,
+      "step": 7120
+    },
+    {
+      "epoch": 4.660130718954249,
+      "grad_norm": 0.9010422229766846,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 7130
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.7102518081665039,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 7140
+    },
+    {
+      "epoch": 4.673202614379085,
+      "grad_norm": 0.7963796257972717,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7150
+    },
+    {
+      "epoch": 4.6797385620915035,
+      "grad_norm": 0.7726007699966431,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 7160
+    },
+    {
+      "epoch": 4.686274509803922,
+      "grad_norm": 0.8097564578056335,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 7170
+    },
+    {
+      "epoch": 4.69281045751634,
+      "grad_norm": 0.9070925116539001,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 7180
+    },
+    {
+      "epoch": 4.699346405228758,
+      "grad_norm": 0.7543528079986572,
+      "learning_rate": 0.0002,
+      "loss": 1.0315,
+      "step": 7190
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.9900904893875122,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 7200
+    },
+    {
+      "epoch": 4.712418300653595,
+      "grad_norm": 0.8033412098884583,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 7210
+    },
+    {
+      "epoch": 4.718954248366013,
+      "grad_norm": 0.8440839052200317,
+      "learning_rate": 0.0002,
+      "loss": 1.1773,
+      "step": 7220
+    },
+    {
+      "epoch": 4.7254901960784315,
+      "grad_norm": 0.9325555562973022,
+      "learning_rate": 0.0002,
+      "loss": 1.1258,
+      "step": 7230
+    },
+    {
+      "epoch": 4.73202614379085,
+      "grad_norm": 0.7881146669387817,
+      "learning_rate": 0.0002,
+      "loss": 1.1384,
+      "step": 7240
+    },
+    {
+      "epoch": 4.738562091503268,
+      "grad_norm": 0.884453296661377,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7250
+    },
+    {
+      "epoch": 4.745098039215686,
+      "grad_norm": 0.9274539351463318,
+      "learning_rate": 0.0002,
+      "loss": 1.1036,
+      "step": 7260
+    },
+    {
+      "epoch": 4.751633986928105,
+      "grad_norm": 1.2367479801177979,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7270
+    },
+    {
+      "epoch": 4.758169934640523,
+      "grad_norm": 0.9499821066856384,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7280
+    },
+    {
+      "epoch": 4.764705882352941,
+      "grad_norm": 2.1918580532073975,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 7290
+    },
+    {
+      "epoch": 4.771241830065359,
+      "grad_norm": 0.8221880793571472,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 7300
+    },
+    {
+      "epoch": 4.777777777777778,
+      "grad_norm": 0.871972918510437,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7310
+    },
+    {
+      "epoch": 4.784313725490196,
+      "grad_norm": 0.8034510612487793,
+      "learning_rate": 0.0002,
+      "loss": 1.0599,
+      "step": 7320
+    },
+    {
+      "epoch": 4.790849673202614,
+      "grad_norm": 0.8959605693817139,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 7330
+    },
+    {
+      "epoch": 4.7973856209150325,
+      "grad_norm": 1.2326215505599976,
+      "learning_rate": 0.0002,
+      "loss": 1.0176,
+      "step": 7340
+    },
+    {
+      "epoch": 4.803921568627451,
+      "grad_norm": 0.9725791811943054,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7350
+    },
+    {
+      "epoch": 4.810457516339869,
+      "grad_norm": 0.7240816354751587,
+      "learning_rate": 0.0002,
+      "loss": 1.1229,
+      "step": 7360
+    },
+    {
+      "epoch": 4.816993464052287,
+      "grad_norm": 0.8265769481658936,
+      "learning_rate": 0.0002,
+      "loss": 1.0669,
+      "step": 7370
+    },
+    {
+      "epoch": 4.823529411764706,
+      "grad_norm": 0.8888696432113647,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 7380
+    },
+    {
+      "epoch": 4.830065359477124,
+      "grad_norm": 0.7776556015014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 7390
+    },
+    {
+      "epoch": 4.836601307189542,
+      "grad_norm": 0.8772371411323547,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7400
+    },
+    {
+      "epoch": 4.8431372549019605,
+      "grad_norm": 0.9786531925201416,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7410
+    },
+    {
+      "epoch": 4.849673202614379,
+      "grad_norm": 0.9059745073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7420
+    },
+    {
+      "epoch": 4.856209150326797,
+      "grad_norm": 0.7422552108764648,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 7430
+    },
+    {
+      "epoch": 4.862745098039216,
+      "grad_norm": 1.3040380477905273,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 7440
+    },
+    {
+      "epoch": 4.8692810457516345,
+      "grad_norm": 1.3278473615646362,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 7450
+    },
+    {
+      "epoch": 4.875816993464053,
+      "grad_norm": 1.2705849409103394,
+      "learning_rate": 0.0002,
+      "loss": 1.0713,
+      "step": 7460
+    },
+    {
+      "epoch": 4.882352941176471,
+      "grad_norm": 0.8837892413139343,
+      "learning_rate": 0.0002,
+      "loss": 1.0034,
+      "step": 7470
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 0.8670691251754761,
+      "learning_rate": 0.0002,
+      "loss": 1.1716,
+      "step": 7480
+    },
+    {
+      "epoch": 4.895424836601308,
+      "grad_norm": 0.9662758111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 7490
+    },
+    {
+      "epoch": 4.901960784313726,
+      "grad_norm": 0.8188302516937256,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 7500
+    },
+    {
+      "epoch": 4.908496732026144,
+      "grad_norm": 0.769442617893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 7510
+    },
+    {
+      "epoch": 4.915032679738562,
+      "grad_norm": 1.1465084552764893,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 7520
+    },
+    {
+      "epoch": 4.921568627450981,
+      "grad_norm": 1.253214955329895,
+      "learning_rate": 0.0002,
+      "loss": 1.0768,
+      "step": 7530
+    },
+    {
+      "epoch": 4.928104575163399,
+      "grad_norm": 0.7922375202178955,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 7540
+    },
+    {
+      "epoch": 4.934640522875817,
+      "grad_norm": 0.8306851387023926,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7550
+    },
+    {
+      "epoch": 4.9411764705882355,
+      "grad_norm": 0.8486151099205017,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 7560
+    },
+    {
+      "epoch": 4.947712418300654,
+      "grad_norm": 1.2601467370986938,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7570
+    },
+    {
+      "epoch": 4.954248366013072,
+      "grad_norm": 0.7980747818946838,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 7580
+    },
+    {
+      "epoch": 4.96078431372549,
+      "grad_norm": 0.8653254508972168,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 7590
+    },
+    {
+      "epoch": 4.967320261437909,
+      "grad_norm": 0.9680571556091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 7600
+    },
+    {
+      "epoch": 4.973856209150327,
+      "grad_norm": 0.9554466605186462,
+      "learning_rate": 0.0002,
+      "loss": 1.1795,
+      "step": 7610
+    },
+    {
+      "epoch": 4.980392156862745,
+      "grad_norm": 1.3693897724151611,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 7620
+    },
+    {
+      "epoch": 4.9869281045751634,
+      "grad_norm": 0.7809282541275024,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 7630
+    },
+    {
+      "epoch": 4.993464052287582,
+      "grad_norm": 0.7528006434440613,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 7640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7491309642791748,
+      "learning_rate": 0.0002,
+      "loss": 0.9951,
+      "step": 7650
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.4197258949279785,
+      "eval_runtime": 33.6327,
+      "eval_samples_per_second": 12.964,
+      "eval_steps_per_second": 1.635,
+      "step": 7650
+    },
+    {
+      "epoch": 5.006535947712418,
+      "grad_norm": 0.8840063214302063,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 7660
+    },
+    {
+      "epoch": 5.0130718954248366,
+      "grad_norm": 1.0118401050567627,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 7670
+    },
+    {
+      "epoch": 5.019607843137255,
+      "grad_norm": 1.0040518045425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1667,
+      "step": 7680
+    },
+    {
+      "epoch": 5.026143790849673,
+      "grad_norm": 0.7541199922561646,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 7690
+    },
+    {
+      "epoch": 5.032679738562091,
+      "grad_norm": 0.9106482863426208,
+      "learning_rate": 0.0002,
+      "loss": 1.0797,
+      "step": 7700
+    },
+    {
+      "epoch": 5.03921568627451,
+      "grad_norm": 1.3691469430923462,
+      "learning_rate": 0.0002,
+      "loss": 1.0096,
+      "step": 7710
+    },
+    {
+      "epoch": 5.045751633986928,
+      "grad_norm": 0.9449689388275146,
+      "learning_rate": 0.0002,
+      "loss": 0.9889,
+      "step": 7720
+    },
+    {
+      "epoch": 5.052287581699346,
+      "grad_norm": 1.1678508520126343,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 7730
+    },
+    {
+      "epoch": 5.0588235294117645,
+      "grad_norm": 1.1296145915985107,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7740
+    },
+    {
+      "epoch": 5.065359477124183,
+      "grad_norm": 0.7863904237747192,
+      "learning_rate": 0.0002,
+      "loss": 0.9339,
+      "step": 7750
+    },
+    {
+      "epoch": 5.071895424836601,
+      "grad_norm": 0.8691433072090149,
+      "learning_rate": 0.0002,
+      "loss": 1.0135,
+      "step": 7760
+    },
+    {
+      "epoch": 5.078431372549019,
+      "grad_norm": 1.0722088813781738,
+      "learning_rate": 0.0002,
+      "loss": 0.9776,
+      "step": 7770
+    },
+    {
+      "epoch": 5.084967320261438,
+      "grad_norm": 0.9625038504600525,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 7780
+    },
+    {
+      "epoch": 5.091503267973856,
+      "grad_norm": 1.2618783712387085,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 7790
+    },
+    {
+      "epoch": 5.098039215686274,
+      "grad_norm": 0.9970650672912598,
+      "learning_rate": 0.0002,
+      "loss": 0.9396,
+      "step": 7800
+    },
+    {
+      "epoch": 5.104575163398692,
+      "grad_norm": 1.3946677446365356,
+      "learning_rate": 0.0002,
+      "loss": 0.9186,
+      "step": 7810
+    },
+    {
+      "epoch": 5.111111111111111,
+      "grad_norm": 1.0260052680969238,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 7820
+    },
+    {
+      "epoch": 5.117647058823529,
+      "grad_norm": 1.105521559715271,
+      "learning_rate": 0.0002,
+      "loss": 0.9865,
+      "step": 7830
+    },
+    {
+      "epoch": 5.124183006535947,
+      "grad_norm": 1.003641128540039,
+      "learning_rate": 0.0002,
+      "loss": 0.9788,
+      "step": 7840
+    },
+    {
+      "epoch": 5.130718954248366,
+      "grad_norm": 1.0315021276474,
+      "learning_rate": 0.0002,
+      "loss": 0.9688,
+      "step": 7850
+    },
+    {
+      "epoch": 5.137254901960785,
+      "grad_norm": 0.9469530582427979,
+      "learning_rate": 0.0002,
+      "loss": 1.0001,
+      "step": 7860
+    },
+    {
+      "epoch": 5.143790849673203,
+      "grad_norm": 1.3244667053222656,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7870
+    },
+    {
+      "epoch": 5.150326797385621,
+      "grad_norm": 1.1732033491134644,
+      "learning_rate": 0.0002,
+      "loss": 0.9657,
+      "step": 7880
+    },
+    {
+      "epoch": 5.1568627450980395,
+      "grad_norm": 1.3129149675369263,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 7890
+    },
+    {
+      "epoch": 5.163398692810458,
+      "grad_norm": 0.8589454293251038,
+      "learning_rate": 0.0002,
+      "loss": 0.9894,
+      "step": 7900
+    },
+    {
+      "epoch": 5.169934640522876,
+      "grad_norm": 0.8954233527183533,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7910
+    },
+    {
+      "epoch": 5.176470588235294,
+      "grad_norm": 0.7426522970199585,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 7920
+    },
+    {
+      "epoch": 5.183006535947713,
+      "grad_norm": 1.1990121603012085,
+      "learning_rate": 0.0002,
+      "loss": 1.0106,
+      "step": 7930
+    },
+    {
+      "epoch": 5.189542483660131,
+      "grad_norm": 0.8867580890655518,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 7940
+    },
+    {
+      "epoch": 5.196078431372549,
+      "grad_norm": 1.016276478767395,
+      "learning_rate": 0.0002,
+      "loss": 0.9727,
+      "step": 7950
+    },
+    {
+      "epoch": 5.2026143790849675,
+      "grad_norm": 1.0210685729980469,
+      "learning_rate": 0.0002,
+      "loss": 0.9908,
+      "step": 7960
+    },
+    {
+      "epoch": 5.209150326797386,
+      "grad_norm": 1.0093122720718384,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7970
+    },
+    {
+      "epoch": 5.215686274509804,
+      "grad_norm": 0.9746801853179932,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 7980
+    },
+    {
+      "epoch": 5.222222222222222,
+      "grad_norm": 0.9113537073135376,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 7990
+    },
+    {
+      "epoch": 5.228758169934641,
+      "grad_norm": 1.2782206535339355,
+      "learning_rate": 0.0002,
+      "loss": 0.9167,
+      "step": 8000
+    },
+    {
+      "epoch": 5.235294117647059,
+      "grad_norm": 1.3223118782043457,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 8010
+    },
+    {
+      "epoch": 5.241830065359477,
+      "grad_norm": 0.7898629307746887,
+      "learning_rate": 0.0002,
+      "loss": 0.9244,
+      "step": 8020
+    },
+    {
+      "epoch": 5.248366013071895,
+      "grad_norm": 0.9822350740432739,
+      "learning_rate": 0.0002,
+      "loss": 1.0574,
+      "step": 8030
+    },
+    {
+      "epoch": 5.254901960784314,
+      "grad_norm": 1.5114340782165527,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 8040
+    },
+    {
+      "epoch": 5.261437908496732,
+      "grad_norm": 0.859006941318512,
+      "learning_rate": 0.0002,
+      "loss": 0.9816,
+      "step": 8050
+    },
+    {
+      "epoch": 5.26797385620915,
+      "grad_norm": 1.0495043992996216,
+      "learning_rate": 0.0002,
+      "loss": 0.9445,
+      "step": 8060
+    },
+    {
+      "epoch": 5.2745098039215685,
+      "grad_norm": 1.329483151435852,
+      "learning_rate": 0.0002,
+      "loss": 0.9724,
+      "step": 8070
+    },
+    {
+      "epoch": 5.281045751633987,
+      "grad_norm": 1.1333061456680298,
+      "learning_rate": 0.0002,
+      "loss": 0.9296,
+      "step": 8080
+    },
+    {
+      "epoch": 5.287581699346405,
+      "grad_norm": 0.8153108358383179,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 8090
+    },
+    {
+      "epoch": 5.294117647058823,
+      "grad_norm": 0.9395004510879517,
+      "learning_rate": 0.0002,
+      "loss": 0.9002,
+      "step": 8100
+    },
+    {
+      "epoch": 5.300653594771242,
+      "grad_norm": 0.8907593488693237,
+      "learning_rate": 0.0002,
+      "loss": 1.0371,
+      "step": 8110
+    },
+    {
+      "epoch": 5.30718954248366,
+      "grad_norm": 0.9808667898178101,
+      "learning_rate": 0.0002,
+      "loss": 0.9301,
+      "step": 8120
+    },
+    {
+      "epoch": 5.313725490196078,
+      "grad_norm": 0.984779417514801,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 8130
+    },
+    {
+      "epoch": 5.3202614379084965,
+      "grad_norm": 0.9787270426750183,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 8140
+    },
+    {
+      "epoch": 5.326797385620915,
+      "grad_norm": 0.9857710599899292,
+      "learning_rate": 0.0002,
+      "loss": 0.9336,
+      "step": 8150
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9774303436279297,
+      "learning_rate": 0.0002,
+      "loss": 0.9884,
+      "step": 8160
+    },
+    {
+      "epoch": 5.339869281045751,
+      "grad_norm": 0.677925169467926,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 8170
+    },
+    {
+      "epoch": 5.34640522875817,
+      "grad_norm": 0.9576456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 8180
+    },
+    {
+      "epoch": 5.352941176470588,
+      "grad_norm": 1.8970937728881836,
+      "learning_rate": 0.0002,
+      "loss": 0.9554,
+      "step": 8190
+    },
+    {
+      "epoch": 5.359477124183006,
+      "grad_norm": 0.9458389282226562,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 8200
+    },
+    {
+      "epoch": 5.366013071895424,
+      "grad_norm": 1.761794924736023,
+      "learning_rate": 0.0002,
+      "loss": 1.0365,
+      "step": 8210
+    },
+    {
+      "epoch": 5.372549019607844,
+      "grad_norm": 1.0693724155426025,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 8220
+    },
+    {
+      "epoch": 5.379084967320262,
+      "grad_norm": 0.9025877714157104,
+      "learning_rate": 0.0002,
+      "loss": 1.0299,
+      "step": 8230
+    },
+    {
+      "epoch": 5.38562091503268,
+      "grad_norm": 1.258857250213623,
+      "learning_rate": 0.0002,
+      "loss": 0.9652,
+      "step": 8240
+    },
+    {
+      "epoch": 5.392156862745098,
+      "grad_norm": 1.084849238395691,
+      "learning_rate": 0.0002,
+      "loss": 0.9735,
+      "step": 8250
+    },
+    {
+      "epoch": 5.398692810457517,
+      "grad_norm": 0.9530340433120728,
+      "learning_rate": 0.0002,
+      "loss": 0.9999,
+      "step": 8260
+    },
+    {
+      "epoch": 5.405228758169935,
+      "grad_norm": 0.830240786075592,
+      "learning_rate": 0.0002,
+      "loss": 1.0268,
+      "step": 8270
+    },
+    {
+      "epoch": 5.411764705882353,
+      "grad_norm": 1.5807015895843506,
+      "learning_rate": 0.0002,
+      "loss": 1.0332,
+      "step": 8280
+    },
+    {
+      "epoch": 5.4183006535947715,
+      "grad_norm": 0.9486905336380005,
+      "learning_rate": 0.0002,
+      "loss": 0.9146,
+      "step": 8290
+    },
+    {
+      "epoch": 5.42483660130719,
+      "grad_norm": 1.0415093898773193,
+      "learning_rate": 0.0002,
+      "loss": 1.0336,
+      "step": 8300
+    },
+    {
+      "epoch": 5.431372549019608,
+      "grad_norm": 1.0501102209091187,
+      "learning_rate": 0.0002,
+      "loss": 0.8933,
+      "step": 8310
+    },
+    {
+      "epoch": 5.437908496732026,
+      "grad_norm": 0.9751836061477661,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 8320
+    },
+    {
+      "epoch": 5.444444444444445,
+      "grad_norm": 1.5529173612594604,
+      "learning_rate": 0.0002,
+      "loss": 1.0755,
+      "step": 8330
+    },
+    {
+      "epoch": 5.450980392156863,
+      "grad_norm": 0.8314350247383118,
+      "learning_rate": 0.0002,
+      "loss": 0.9814,
+      "step": 8340
+    },
+    {
+      "epoch": 5.457516339869281,
+      "grad_norm": 1.2555103302001953,
+      "learning_rate": 0.0002,
+      "loss": 1.0596,
+      "step": 8350
+    },
+    {
+      "epoch": 5.4640522875816995,
+      "grad_norm": 0.9408367872238159,
+      "learning_rate": 0.0002,
+      "loss": 1.0127,
+      "step": 8360
+    },
+    {
+      "epoch": 5.470588235294118,
+      "grad_norm": 0.9483312964439392,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 8370
+    },
+    {
+      "epoch": 5.477124183006536,
+      "grad_norm": 0.957905650138855,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8380
+    },
+    {
+      "epoch": 5.483660130718954,
+      "grad_norm": 1.4000147581100464,
+      "learning_rate": 0.0002,
+      "loss": 1.0985,
+      "step": 8390
+    },
+    {
+      "epoch": 5.490196078431373,
+      "grad_norm": 1.7032461166381836,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 8400
+    },
+    {
+      "epoch": 5.496732026143791,
+      "grad_norm": 0.8978716731071472,
+      "learning_rate": 0.0002,
+      "loss": 0.9539,
+      "step": 8410
+    },
+    {
+      "epoch": 5.503267973856209,
+      "grad_norm": 0.8659300804138184,
+      "learning_rate": 0.0002,
+      "loss": 0.9544,
+      "step": 8420
+    },
+    {
+      "epoch": 5.509803921568627,
+      "grad_norm": 1.3629727363586426,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 8430
+    },
+    {
+      "epoch": 5.516339869281046,
+      "grad_norm": 1.2741984128952026,
+      "learning_rate": 0.0002,
+      "loss": 0.9696,
+      "step": 8440
+    },
+    {
+      "epoch": 5.522875816993464,
+      "grad_norm": 1.3867180347442627,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 8450
+    },
+    {
+      "epoch": 5.529411764705882,
+      "grad_norm": 1.0662001371383667,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 8460
+    },
+    {
+      "epoch": 5.5359477124183005,
+      "grad_norm": 1.7005380392074585,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 8470
+    },
+    {
+      "epoch": 5.542483660130719,
+      "grad_norm": 1.3730385303497314,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 8480
+    },
+    {
+      "epoch": 5.549019607843137,
+      "grad_norm": 1.7737441062927246,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 8490
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.907487690448761,
+      "learning_rate": 0.0002,
+      "loss": 0.9729,
+      "step": 8500
+    },
+    {
+      "epoch": 5.562091503267974,
+      "grad_norm": 0.8882441520690918,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 8510
+    },
+    {
+      "epoch": 5.568627450980392,
+      "grad_norm": 0.8655388951301575,
+      "learning_rate": 0.0002,
+      "loss": 0.973,
+      "step": 8520
+    },
+    {
+      "epoch": 5.57516339869281,
+      "grad_norm": 1.379992961883545,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 8530
+    },
+    {
+      "epoch": 5.5816993464052285,
+      "grad_norm": 1.0021201372146606,
+      "learning_rate": 0.0002,
+      "loss": 1.0174,
+      "step": 8540
+    },
+    {
+      "epoch": 5.588235294117647,
+      "grad_norm": 1.2636926174163818,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 8550
+    },
+    {
+      "epoch": 5.594771241830065,
+      "grad_norm": 1.279025912284851,
+      "learning_rate": 0.0002,
+      "loss": 1.0243,
+      "step": 8560
+    },
+    {
+      "epoch": 5.601307189542483,
+      "grad_norm": 0.8885834217071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9917,
+      "step": 8570
+    },
+    {
+      "epoch": 5.607843137254902,
+      "grad_norm": 1.1975032091140747,
+      "learning_rate": 0.0002,
+      "loss": 0.9849,
+      "step": 8580
+    },
+    {
+      "epoch": 5.61437908496732,
+      "grad_norm": 1.005470871925354,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8590
+    },
+    {
+      "epoch": 5.620915032679738,
+      "grad_norm": 1.104286551475525,
+      "learning_rate": 0.0002,
+      "loss": 0.9947,
+      "step": 8600
+    },
+    {
+      "epoch": 5.627450980392156,
+      "grad_norm": 1.435445785522461,
+      "learning_rate": 0.0002,
+      "loss": 1.0585,
+      "step": 8610
+    },
+    {
+      "epoch": 5.633986928104575,
+      "grad_norm": 1.0270172357559204,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 8620
+    },
+    {
+      "epoch": 5.640522875816993,
+      "grad_norm": 1.0929527282714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 8630
+    },
+    {
+      "epoch": 5.647058823529412,
+      "grad_norm": 1.1061221361160278,
+      "learning_rate": 0.0002,
+      "loss": 0.9694,
+      "step": 8640
+    },
+    {
+      "epoch": 5.65359477124183,
+      "grad_norm": 0.9563149213790894,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 8650
+    },
+    {
+      "epoch": 5.660130718954249,
+      "grad_norm": 1.0434954166412354,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 8660
+    },
+    {
+      "epoch": 5.666666666666667,
+      "grad_norm": 1.3695117235183716,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 8670
+    },
+    {
+      "epoch": 5.673202614379085,
+      "grad_norm": 1.0540564060211182,
+      "learning_rate": 0.0002,
+      "loss": 0.9441,
+      "step": 8680
+    },
+    {
+      "epoch": 5.6797385620915035,
+      "grad_norm": 1.5942492485046387,
+      "learning_rate": 0.0002,
+      "loss": 0.9755,
+      "step": 8690
+    },
+    {
+      "epoch": 5.686274509803922,
+      "grad_norm": 0.9485495090484619,
+      "learning_rate": 0.0002,
+      "loss": 1.0071,
+      "step": 8700
+    },
+    {
+      "epoch": 5.69281045751634,
+      "grad_norm": 1.1483162641525269,
+      "learning_rate": 0.0002,
+      "loss": 0.9998,
+      "step": 8710
+    },
+    {
+      "epoch": 5.699346405228758,
+      "grad_norm": 0.9075471758842468,
+      "learning_rate": 0.0002,
+      "loss": 0.9578,
+      "step": 8720
+    },
+    {
+      "epoch": 5.705882352941177,
+      "grad_norm": 1.7908551692962646,
+      "learning_rate": 0.0002,
+      "loss": 0.9488,
+      "step": 8730
+    },
+    {
+      "epoch": 5.712418300653595,
+      "grad_norm": 0.8867162466049194,
+      "learning_rate": 0.0002,
+      "loss": 1.0163,
+      "step": 8740
+    },
+    {
+      "epoch": 5.718954248366013,
+      "grad_norm": 1.7165148258209229,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 8750
+    },
+    {
+      "epoch": 5.7254901960784315,
+      "grad_norm": 0.9529356956481934,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 8760
+    },
+    {
+      "epoch": 5.73202614379085,
+      "grad_norm": 1.01852548122406,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 8770
+    },
+    {
+      "epoch": 5.738562091503268,
+      "grad_norm": 0.9538423418998718,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 8780
+    },
+    {
+      "epoch": 5.745098039215686,
+      "grad_norm": 0.9007737636566162,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 8790
+    },
+    {
+      "epoch": 5.751633986928105,
+      "grad_norm": 0.9107874035835266,
+      "learning_rate": 0.0002,
+      "loss": 0.9766,
+      "step": 8800
+    },
+    {
+      "epoch": 5.758169934640523,
+      "grad_norm": 0.7379238605499268,
+      "learning_rate": 0.0002,
+      "loss": 0.9212,
+      "step": 8810
+    },
+    {
+      "epoch": 5.764705882352941,
+      "grad_norm": 1.072645902633667,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 8820
+    },
+    {
+      "epoch": 5.771241830065359,
+      "grad_norm": 1.002008080482483,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 8830
+    },
+    {
+      "epoch": 5.777777777777778,
+      "grad_norm": 1.0435924530029297,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 8840
+    },
+    {
+      "epoch": 5.784313725490196,
+      "grad_norm": 0.9874551296234131,
+      "learning_rate": 0.0002,
+      "loss": 0.9458,
+      "step": 8850
+    },
+    {
+      "epoch": 5.790849673202614,
+      "grad_norm": 1.1729662418365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1241,
+      "step": 8860
+    },
+    {
+      "epoch": 5.7973856209150325,
+      "grad_norm": 1.3300775289535522,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 8870
+    },
+    {
+      "epoch": 5.803921568627451,
+      "grad_norm": 1.612707257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.0989,
+      "step": 8880
+    },
+    {
+      "epoch": 5.810457516339869,
+      "grad_norm": 0.9047797322273254,
+      "learning_rate": 0.0002,
+      "loss": 0.9119,
+      "step": 8890
+    },
+    {
+      "epoch": 5.816993464052287,
+      "grad_norm": 1.0958741903305054,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 8900
+    },
+    {
+      "epoch": 5.823529411764706,
+      "grad_norm": 1.0099612474441528,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 8910
+    },
+    {
+      "epoch": 5.830065359477124,
+      "grad_norm": 0.8442328572273254,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 8920
+    },
+    {
+      "epoch": 5.836601307189542,
+      "grad_norm": 1.1388301849365234,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 8930
+    },
+    {
+      "epoch": 5.8431372549019605,
+      "grad_norm": 0.8296026587486267,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 8940
+    },
+    {
+      "epoch": 5.849673202614379,
+      "grad_norm": 1.0843533277511597,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8950
+    },
+    {
+      "epoch": 5.856209150326797,
+      "grad_norm": 0.8496834635734558,
+      "learning_rate": 0.0002,
+      "loss": 1.0009,
+      "step": 8960
+    },
+    {
+      "epoch": 5.862745098039216,
+      "grad_norm": 1.6894690990447998,
+      "learning_rate": 0.0002,
+      "loss": 0.9927,
+      "step": 8970
+    },
+    {
+      "epoch": 5.8692810457516345,
+      "grad_norm": 1.0012282133102417,
+      "learning_rate": 0.0002,
+      "loss": 1.0939,
+      "step": 8980
+    },
+    {
+      "epoch": 5.875816993464053,
+      "grad_norm": 0.8521103262901306,
+      "learning_rate": 0.0002,
+      "loss": 0.9722,
+      "step": 8990
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 1.246841311454773,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 9000
+    },
+    {
+      "epoch": 5.888888888888889,
+      "grad_norm": 0.9941892027854919,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 9010
+    },
+    {
+      "epoch": 5.895424836601308,
+      "grad_norm": 1.067413568496704,
+      "learning_rate": 0.0002,
+      "loss": 0.8754,
+      "step": 9020
+    },
+    {
+      "epoch": 5.901960784313726,
+      "grad_norm": 1.0045088529586792,
+      "learning_rate": 0.0002,
+      "loss": 1.0153,
+      "step": 9030
+    },
+    {
+      "epoch": 5.908496732026144,
+      "grad_norm": 1.383063554763794,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 9040
+    },
+    {
+      "epoch": 5.915032679738562,
+      "grad_norm": 0.8754428625106812,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 9050
+    },
+    {
+      "epoch": 5.921568627450981,
+      "grad_norm": 0.8577388525009155,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 9060
+    },
+    {
+      "epoch": 5.928104575163399,
+      "grad_norm": 0.8718975186347961,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 9070
+    },
+    {
+      "epoch": 5.934640522875817,
+      "grad_norm": 1.1762131452560425,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 9080
+    },
+    {
+      "epoch": 5.9411764705882355,
+      "grad_norm": 1.1025866270065308,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 9090
+    },
+    {
+      "epoch": 5.947712418300654,
+      "grad_norm": 1.0439870357513428,
+      "learning_rate": 0.0002,
+      "loss": 0.9155,
+      "step": 9100
+    },
+    {
+      "epoch": 5.954248366013072,
+      "grad_norm": 1.2411525249481201,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 9110
+    },
+    {
+      "epoch": 5.96078431372549,
+      "grad_norm": 1.0317714214324951,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 9120
+    },
+    {
+      "epoch": 5.967320261437909,
+      "grad_norm": 0.9880492091178894,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 9130
+    },
+    {
+      "epoch": 5.973856209150327,
+      "grad_norm": 0.9039815664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 9140
+    },
+    {
+      "epoch": 5.980392156862745,
+      "grad_norm": 0.9049116373062134,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 9150
+    },
+    {
+      "epoch": 5.9869281045751634,
+      "grad_norm": 0.996749222278595,
+      "learning_rate": 0.0002,
+      "loss": 0.9792,
+      "step": 9160
+    },
+    {
+      "epoch": 5.993464052287582,
+      "grad_norm": 0.8716062307357788,
+      "learning_rate": 0.0002,
+      "loss": 0.8857,
+      "step": 9170
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.3081294298171997,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 9180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.45111083984375,
+      "eval_runtime": 34.7121,
+      "eval_samples_per_second": 12.56,
+      "eval_steps_per_second": 1.584,
+      "step": 9180
+    },
+    {
+      "epoch": 6.006535947712418,
+      "grad_norm": 1.1378029584884644,
+      "learning_rate": 0.0002,
+      "loss": 0.9306,
+      "step": 9190
+    },
+    {
+      "epoch": 6.0130718954248366,
+      "grad_norm": 1.2921233177185059,
+      "learning_rate": 0.0002,
+      "loss": 0.8794,
+      "step": 9200
+    },
+    {
+      "epoch": 6.019607843137255,
+      "grad_norm": 1.039211630821228,
+      "learning_rate": 0.0002,
+      "loss": 0.8145,
+      "step": 9210
+    },
+    {
+      "epoch": 6.026143790849673,
+      "grad_norm": 0.9715196490287781,
+      "learning_rate": 0.0002,
+      "loss": 0.8524,
+      "step": 9220
+    },
+    {
+      "epoch": 6.032679738562091,
+      "grad_norm": 1.220642328262329,
+      "learning_rate": 0.0002,
+      "loss": 1.035,
+      "step": 9230
+    },
+    {
+      "epoch": 6.03921568627451,
+      "grad_norm": 0.854360044002533,
+      "learning_rate": 0.0002,
+      "loss": 0.8468,
+      "step": 9240
+    },
+    {
+      "epoch": 6.045751633986928,
+      "grad_norm": 0.8806933164596558,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 9250
+    },
+    {
+      "epoch": 6.052287581699346,
+      "grad_norm": 1.4315874576568604,
+      "learning_rate": 0.0002,
+      "loss": 0.8305,
+      "step": 9260
+    },
+    {
+      "epoch": 6.0588235294117645,
+      "grad_norm": 0.9382007122039795,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 9270
+    },
+    {
+      "epoch": 6.065359477124183,
+      "grad_norm": 1.2184561491012573,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 9280
+    },
+    {
+      "epoch": 6.071895424836601,
+      "grad_norm": 1.2331548929214478,
+      "learning_rate": 0.0002,
+      "loss": 0.8806,
+      "step": 9290
+    },
+    {
+      "epoch": 6.078431372549019,
+      "grad_norm": 1.1112796068191528,
+      "learning_rate": 0.0002,
+      "loss": 0.8354,
+      "step": 9300
+    },
+    {
+      "epoch": 6.084967320261438,
+      "grad_norm": 1.4753731489181519,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 9310
+    },
+    {
+      "epoch": 6.091503267973856,
+      "grad_norm": 1.2783401012420654,
+      "learning_rate": 0.0002,
+      "loss": 0.9198,
+      "step": 9320
+    },
+    {
+      "epoch": 6.098039215686274,
+      "grad_norm": 0.9916909337043762,
+      "learning_rate": 0.0002,
+      "loss": 0.8294,
+      "step": 9330
+    },
+    {
+      "epoch": 6.104575163398692,
+      "grad_norm": 0.9300099015235901,
+      "learning_rate": 0.0002,
+      "loss": 0.876,
+      "step": 9340
+    },
+    {
+      "epoch": 6.111111111111111,
+      "grad_norm": 1.4985264539718628,
+      "learning_rate": 0.0002,
+      "loss": 0.9064,
+      "step": 9350
+    },
+    {
+      "epoch": 6.117647058823529,
+      "grad_norm": 1.276380181312561,
+      "learning_rate": 0.0002,
+      "loss": 1.0106,
+      "step": 9360
+    },
+    {
+      "epoch": 6.124183006535947,
+      "grad_norm": 1.181113600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9068,
+      "step": 9370
+    },
+    {
+      "epoch": 6.130718954248366,
+      "grad_norm": 1.698729395866394,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 9380
+    },
+    {
+      "epoch": 6.137254901960785,
+      "grad_norm": 0.9793189764022827,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 9390
+    },
+    {
+      "epoch": 6.143790849673203,
+      "grad_norm": 1.1942132711410522,
+      "learning_rate": 0.0002,
+      "loss": 0.9731,
+      "step": 9400
+    },
+    {
+      "epoch": 6.150326797385621,
+      "grad_norm": 1.2160184383392334,
+      "learning_rate": 0.0002,
+      "loss": 0.8762,
+      "step": 9410
+    },
+    {
+      "epoch": 6.1568627450980395,
+      "grad_norm": 1.0802825689315796,
+      "learning_rate": 0.0002,
+      "loss": 0.801,
+      "step": 9420
+    },
+    {
+      "epoch": 6.163398692810458,
+      "grad_norm": 3.024529218673706,
+      "learning_rate": 0.0002,
+      "loss": 0.9055,
+      "step": 9430
+    },
+    {
+      "epoch": 6.169934640522876,
+      "grad_norm": 0.975062370300293,
+      "learning_rate": 0.0002,
+      "loss": 0.8739,
+      "step": 9440
+    },
+    {
+      "epoch": 6.176470588235294,
+      "grad_norm": 0.9243306517601013,
+      "learning_rate": 0.0002,
+      "loss": 0.8485,
+      "step": 9450
+    },
+    {
+      "epoch": 6.183006535947713,
+      "grad_norm": 0.8892099857330322,
+      "learning_rate": 0.0002,
+      "loss": 0.947,
+      "step": 9460
+    },
+    {
+      "epoch": 6.189542483660131,
+      "grad_norm": 1.4151731729507446,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 9470
+    },
+    {
+      "epoch": 6.196078431372549,
+      "grad_norm": 1.064701795578003,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 9480
+    },
+    {
+      "epoch": 6.2026143790849675,
+      "grad_norm": 1.1104519367218018,
+      "learning_rate": 0.0002,
+      "loss": 0.906,
+      "step": 9490
+    },
+    {
+      "epoch": 6.209150326797386,
+      "grad_norm": 1.4788947105407715,
+      "learning_rate": 0.0002,
+      "loss": 0.9572,
+      "step": 9500
+    },
+    {
+      "epoch": 6.215686274509804,
+      "grad_norm": 0.7976077795028687,
+      "learning_rate": 0.0002,
+      "loss": 0.8014,
+      "step": 9510
+    },
+    {
+      "epoch": 6.222222222222222,
+      "grad_norm": 1.256864070892334,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 9520
+    },
+    {
+      "epoch": 6.228758169934641,
+      "grad_norm": 1.3874554634094238,
+      "learning_rate": 0.0002,
+      "loss": 0.9104,
+      "step": 9530
+    },
+    {
+      "epoch": 6.235294117647059,
+      "grad_norm": 1.9012963771820068,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 9540
+    },
+    {
+      "epoch": 6.241830065359477,
+      "grad_norm": 1.275212287902832,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 9550
+    },
+    {
+      "epoch": 6.248366013071895,
+      "grad_norm": 1.1007417440414429,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 9560
+    },
+    {
+      "epoch": 6.254901960784314,
+      "grad_norm": 1.0602147579193115,
+      "learning_rate": 0.0002,
+      "loss": 0.9191,
+      "step": 9570
+    },
+    {
+      "epoch": 6.261437908496732,
+      "grad_norm": 1.2276418209075928,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 9580
+    },
+    {
+      "epoch": 6.26797385620915,
+      "grad_norm": 1.0111924409866333,
+      "learning_rate": 0.0002,
+      "loss": 0.9363,
+      "step": 9590
+    },
+    {
+      "epoch": 6.2745098039215685,
+      "grad_norm": 0.9031485915184021,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 9600
+    },
+    {
+      "epoch": 6.281045751633987,
+      "grad_norm": 0.9893783926963806,
+      "learning_rate": 0.0002,
+      "loss": 0.9138,
+      "step": 9610
+    },
+    {
+      "epoch": 6.287581699346405,
+      "grad_norm": 1.1979725360870361,
+      "learning_rate": 0.0002,
+      "loss": 0.9114,
+      "step": 9620
+    },
+    {
+      "epoch": 6.294117647058823,
+      "grad_norm": 1.380516767501831,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 9630
+    },
+    {
+      "epoch": 6.300653594771242,
+      "grad_norm": 1.1370083093643188,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 9640
+    },
+    {
+      "epoch": 6.30718954248366,
+      "grad_norm": 1.4091558456420898,
+      "learning_rate": 0.0002,
+      "loss": 0.9073,
+      "step": 9650
+    },
+    {
+      "epoch": 6.313725490196078,
+      "grad_norm": 1.0670944452285767,
+      "learning_rate": 0.0002,
+      "loss": 0.9096,
+      "step": 9660
+    },
+    {
+      "epoch": 6.3202614379084965,
+      "grad_norm": 0.9150263667106628,
+      "learning_rate": 0.0002,
+      "loss": 0.9376,
+      "step": 9670
+    },
+    {
+      "epoch": 6.326797385620915,
+      "grad_norm": 1.1342853307724,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 9680
+    },
+    {
+      "epoch": 6.333333333333333,
+      "grad_norm": 1.2733415365219116,
+      "learning_rate": 0.0002,
+      "loss": 1.002,
+      "step": 9690
+    },
+    {
+      "epoch": 6.339869281045751,
+      "grad_norm": 1.3647292852401733,
+      "learning_rate": 0.0002,
+      "loss": 0.9579,
+      "step": 9700
+    },
+    {
+      "epoch": 6.34640522875817,
+      "grad_norm": 1.0435094833374023,
+      "learning_rate": 0.0002,
+      "loss": 0.87,
+      "step": 9710
+    },
+    {
+      "epoch": 6.352941176470588,
+      "grad_norm": 1.3641071319580078,
+      "learning_rate": 0.0002,
+      "loss": 0.8812,
+      "step": 9720
+    },
+    {
+      "epoch": 6.359477124183006,
+      "grad_norm": 1.2806159257888794,
+      "learning_rate": 0.0002,
+      "loss": 0.8888,
+      "step": 9730
+    },
+    {
+      "epoch": 6.366013071895424,
+      "grad_norm": 1.0193076133728027,
+      "learning_rate": 0.0002,
+      "loss": 0.9481,
+      "step": 9740
+    },
+    {
+      "epoch": 6.372549019607844,
+      "grad_norm": 1.2349408864974976,
+      "learning_rate": 0.0002,
+      "loss": 0.931,
+      "step": 9750
+    },
+    {
+      "epoch": 6.379084967320262,
+      "grad_norm": 1.2062549591064453,
+      "learning_rate": 0.0002,
+      "loss": 0.8837,
+      "step": 9760
+    },
+    {
+      "epoch": 6.38562091503268,
+      "grad_norm": 1.4402194023132324,
+      "learning_rate": 0.0002,
+      "loss": 0.8947,
+      "step": 9770
+    },
+    {
+      "epoch": 6.392156862745098,
+      "grad_norm": 1.1730891466140747,
+      "learning_rate": 0.0002,
+      "loss": 0.8724,
+      "step": 9780
+    },
+    {
+      "epoch": 6.398692810457517,
+      "grad_norm": 1.1481093168258667,
+      "learning_rate": 0.0002,
+      "loss": 0.9005,
+      "step": 9790
+    },
+    {
+      "epoch": 6.405228758169935,
+      "grad_norm": 1.0012723207473755,
+      "learning_rate": 0.0002,
+      "loss": 0.9431,
+      "step": 9800
+    },
+    {
+      "epoch": 6.411764705882353,
+      "grad_norm": 0.8839848041534424,
+      "learning_rate": 0.0002,
+      "loss": 0.8856,
+      "step": 9810
+    },
+    {
+      "epoch": 6.4183006535947715,
+      "grad_norm": 1.096693992614746,
+      "learning_rate": 0.0002,
+      "loss": 0.8147,
+      "step": 9820
+    },
+    {
+      "epoch": 6.42483660130719,
+      "grad_norm": 1.4713369607925415,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 9830
+    },
+    {
+      "epoch": 6.431372549019608,
+      "grad_norm": 1.2529761791229248,
+      "learning_rate": 0.0002,
+      "loss": 0.9563,
+      "step": 9840
+    },
+    {
+      "epoch": 6.437908496732026,
+      "grad_norm": 1.5575600862503052,
+      "learning_rate": 0.0002,
+      "loss": 0.8551,
+      "step": 9850
+    },
+    {
+      "epoch": 6.444444444444445,
+      "grad_norm": 1.2188916206359863,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 9860
+    },
+    {
+      "epoch": 6.450980392156863,
+      "grad_norm": 1.1558794975280762,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 9870
+    },
+    {
+      "epoch": 6.457516339869281,
+      "grad_norm": 1.1506937742233276,
+      "learning_rate": 0.0002,
+      "loss": 0.8632,
+      "step": 9880
+    },
+    {
+      "epoch": 6.4640522875816995,
+      "grad_norm": 1.1168335676193237,
+      "learning_rate": 0.0002,
+      "loss": 1.0575,
+      "step": 9890
+    },
+    {
+      "epoch": 6.470588235294118,
+      "grad_norm": 1.192449688911438,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 9900
+    },
+    {
+      "epoch": 6.477124183006536,
+      "grad_norm": 1.0451104640960693,
+      "learning_rate": 0.0002,
+      "loss": 0.9478,
+      "step": 9910
+    },
+    {
+      "epoch": 6.483660130718954,
+      "grad_norm": 1.1111775636672974,
+      "learning_rate": 0.0002,
+      "loss": 0.9034,
+      "step": 9920
+    },
+    {
+      "epoch": 6.490196078431373,
+      "grad_norm": 1.2094531059265137,
+      "learning_rate": 0.0002,
+      "loss": 0.8971,
+      "step": 9930
+    },
+    {
+      "epoch": 6.496732026143791,
+      "grad_norm": 1.0547380447387695,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 9940
+    },
+    {
+      "epoch": 6.503267973856209,
+      "grad_norm": 1.5547202825546265,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 9950
+    },
+    {
+      "epoch": 6.509803921568627,
+      "grad_norm": 1.1917903423309326,
+      "learning_rate": 0.0002,
+      "loss": 0.9109,
+      "step": 9960
+    },
+    {
+      "epoch": 6.516339869281046,
+      "grad_norm": 1.0918153524398804,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 9970
+    },
+    {
+      "epoch": 6.522875816993464,
+      "grad_norm": 1.146968960762024,
+      "learning_rate": 0.0002,
+      "loss": 0.8752,
+      "step": 9980
+    },
+    {
+      "epoch": 6.529411764705882,
+      "grad_norm": 0.9899234771728516,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 9990
+    },
+    {
+      "epoch": 6.5359477124183005,
+      "grad_norm": 2.160924196243286,
+      "learning_rate": 0.0002,
+      "loss": 0.91,
+      "step": 10000
+    },
+    {
+      "epoch": 6.542483660130719,
+      "grad_norm": 1.6366891860961914,
+      "learning_rate": 0.0002,
+      "loss": 0.9683,
+      "step": 10010
+    },
+    {
+      "epoch": 6.549019607843137,
+      "grad_norm": 0.9876762628555298,
+      "learning_rate": 0.0002,
+      "loss": 0.8582,
+      "step": 10020
+    },
+    {
+      "epoch": 6.555555555555555,
+      "grad_norm": 1.5622549057006836,
+      "learning_rate": 0.0002,
+      "loss": 0.8385,
+      "step": 10030
+    },
+    {
+      "epoch": 6.562091503267974,
+      "grad_norm": 1.0108020305633545,
+      "learning_rate": 0.0002,
+      "loss": 0.8791,
+      "step": 10040
+    },
+    {
+      "epoch": 6.568627450980392,
+      "grad_norm": 1.0725725889205933,
+      "learning_rate": 0.0002,
+      "loss": 0.9574,
+      "step": 10050
+    },
+    {
+      "epoch": 6.57516339869281,
+      "grad_norm": 1.1551216840744019,
+      "learning_rate": 0.0002,
+      "loss": 0.8297,
+      "step": 10060
+    },
+    {
+      "epoch": 6.5816993464052285,
+      "grad_norm": 1.5174646377563477,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 10070
+    },
+    {
+      "epoch": 6.588235294117647,
+      "grad_norm": 1.041877031326294,
+      "learning_rate": 0.0002,
+      "loss": 0.8203,
+      "step": 10080
+    },
+    {
+      "epoch": 6.594771241830065,
+      "grad_norm": 0.9939621686935425,
+      "learning_rate": 0.0002,
+      "loss": 0.9684,
+      "step": 10090
+    },
+    {
+      "epoch": 6.601307189542483,
+      "grad_norm": 1.2706589698791504,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 10100
+    },
+    {
+      "epoch": 6.607843137254902,
+      "grad_norm": 1.1071467399597168,
+      "learning_rate": 0.0002,
+      "loss": 0.9614,
+      "step": 10110
+    },
+    {
+      "epoch": 6.61437908496732,
+      "grad_norm": 0.9449541568756104,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 10120
+    },
+    {
+      "epoch": 6.620915032679738,
+      "grad_norm": 1.0961830615997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9557,
+      "step": 10130
+    },
+    {
+      "epoch": 6.627450980392156,
+      "grad_norm": 1.7726300954818726,
+      "learning_rate": 0.0002,
+      "loss": 0.9865,
+      "step": 10140
+    },
+    {
+      "epoch": 6.633986928104575,
+      "grad_norm": 1.2345516681671143,
+      "learning_rate": 0.0002,
+      "loss": 0.9657,
+      "step": 10150
+    },
+    {
+      "epoch": 6.640522875816993,
+      "grad_norm": 1.2062907218933105,
+      "learning_rate": 0.0002,
+      "loss": 0.9573,
+      "step": 10160
+    },
+    {
+      "epoch": 6.647058823529412,
+      "grad_norm": 1.029327154159546,
+      "learning_rate": 0.0002,
+      "loss": 0.918,
+      "step": 10170
+    },
+    {
+      "epoch": 6.65359477124183,
+      "grad_norm": 1.442307710647583,
+      "learning_rate": 0.0002,
+      "loss": 0.9211,
+      "step": 10180
+    },
+    {
+      "epoch": 6.660130718954249,
+      "grad_norm": 1.2579066753387451,
+      "learning_rate": 0.0002,
+      "loss": 0.8924,
+      "step": 10190
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.4563188552856445,
+      "learning_rate": 0.0002,
+      "loss": 0.9836,
+      "step": 10200
+    },
+    {
+      "epoch": 6.673202614379085,
+      "grad_norm": 0.9699450135231018,
+      "learning_rate": 0.0002,
+      "loss": 0.8876,
+      "step": 10210
+    },
+    {
+      "epoch": 6.6797385620915035,
+      "grad_norm": 1.812523603439331,
+      "learning_rate": 0.0002,
+      "loss": 0.9589,
+      "step": 10220
+    },
+    {
+      "epoch": 6.686274509803922,
+      "grad_norm": 1.124000906944275,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 10230
+    },
+    {
+      "epoch": 6.69281045751634,
+      "grad_norm": 1.0957475900650024,
+      "learning_rate": 0.0002,
+      "loss": 0.8924,
+      "step": 10240
+    },
+    {
+      "epoch": 6.699346405228758,
+      "grad_norm": 0.989689826965332,
+      "learning_rate": 0.0002,
+      "loss": 0.8891,
+      "step": 10250
+    },
+    {
+      "epoch": 6.705882352941177,
+      "grad_norm": 1.4353317022323608,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 10260
+    },
+    {
+      "epoch": 6.712418300653595,
+      "grad_norm": 1.0245451927185059,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 10270
+    },
+    {
+      "epoch": 6.718954248366013,
+      "grad_norm": 1.097334861755371,
+      "learning_rate": 0.0002,
+      "loss": 0.8814,
+      "step": 10280
+    },
+    {
+      "epoch": 6.7254901960784315,
+      "grad_norm": 0.982356071472168,
+      "learning_rate": 0.0002,
+      "loss": 0.9927,
+      "step": 10290
+    },
+    {
+      "epoch": 6.73202614379085,
+      "grad_norm": 1.8842819929122925,
+      "learning_rate": 0.0002,
+      "loss": 0.9909,
+      "step": 10300
+    },
+    {
+      "epoch": 6.738562091503268,
+      "grad_norm": 0.8648947477340698,
+      "learning_rate": 0.0002,
+      "loss": 0.9286,
+      "step": 10310
+    },
+    {
+      "epoch": 6.745098039215686,
+      "grad_norm": 1.1510577201843262,
+      "learning_rate": 0.0002,
+      "loss": 0.987,
+      "step": 10320
+    },
+    {
+      "epoch": 6.751633986928105,
+      "grad_norm": 1.874495506286621,
+      "learning_rate": 0.0002,
+      "loss": 0.9217,
+      "step": 10330
+    },
+    {
+      "epoch": 6.758169934640523,
+      "grad_norm": 1.1126408576965332,
+      "learning_rate": 0.0002,
+      "loss": 0.8914,
+      "step": 10340
+    },
+    {
+      "epoch": 6.764705882352941,
+      "grad_norm": 1.6654644012451172,
+      "learning_rate": 0.0002,
+      "loss": 0.8508,
+      "step": 10350
+    },
+    {
+      "epoch": 6.771241830065359,
+      "grad_norm": 1.0699580907821655,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 10360
+    },
+    {
+      "epoch": 6.777777777777778,
+      "grad_norm": 0.9460757374763489,
+      "learning_rate": 0.0002,
+      "loss": 0.882,
+      "step": 10370
+    },
+    {
+      "epoch": 6.784313725490196,
+      "grad_norm": 1.2553058862686157,
+      "learning_rate": 0.0002,
+      "loss": 0.9589,
+      "step": 10380
+    },
+    {
+      "epoch": 6.790849673202614,
+      "grad_norm": 1.0939891338348389,
+      "learning_rate": 0.0002,
+      "loss": 0.8782,
+      "step": 10390
+    },
+    {
+      "epoch": 6.7973856209150325,
+      "grad_norm": 1.0647451877593994,
+      "learning_rate": 0.0002,
+      "loss": 0.9189,
+      "step": 10400
+    },
+    {
+      "epoch": 6.803921568627451,
+      "grad_norm": 1.0954521894454956,
+      "learning_rate": 0.0002,
+      "loss": 0.9478,
+      "step": 10410
+    },
+    {
+      "epoch": 6.810457516339869,
+      "grad_norm": 1.4371392726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.0385,
+      "step": 10420
+    },
+    {
+      "epoch": 6.816993464052287,
+      "grad_norm": 1.0063464641571045,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 10430
+    },
+    {
+      "epoch": 6.823529411764706,
+      "grad_norm": 1.5189263820648193,
+      "learning_rate": 0.0002,
+      "loss": 0.8737,
+      "step": 10440
+    },
+    {
+      "epoch": 6.830065359477124,
+      "grad_norm": 0.9715501070022583,
+      "learning_rate": 0.0002,
+      "loss": 0.9246,
+      "step": 10450
+    },
+    {
+      "epoch": 6.836601307189542,
+      "grad_norm": 1.114586353302002,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 10460
+    },
+    {
+      "epoch": 6.8431372549019605,
+      "grad_norm": 1.2991431951522827,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 10470
+    },
+    {
+      "epoch": 6.849673202614379,
+      "grad_norm": 1.203114628791809,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 10480
+    },
+    {
+      "epoch": 6.856209150326797,
+      "grad_norm": 1.476167917251587,
+      "learning_rate": 0.0002,
+      "loss": 1.0032,
+      "step": 10490
+    },
+    {
+      "epoch": 6.862745098039216,
+      "grad_norm": 1.0933326482772827,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 10500
+    },
+    {
+      "epoch": 6.8692810457516345,
+      "grad_norm": 1.2831504344940186,
+      "learning_rate": 0.0002,
+      "loss": 1.0068,
+      "step": 10510
+    },
+    {
+      "epoch": 6.875816993464053,
+      "grad_norm": 1.1967637538909912,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 10520
+    },
+    {
+      "epoch": 6.882352941176471,
+      "grad_norm": 1.1276888847351074,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 10530
+    },
+    {
+      "epoch": 6.888888888888889,
+      "grad_norm": 1.2680490016937256,
+      "learning_rate": 0.0002,
+      "loss": 0.9568,
+      "step": 10540
+    },
+    {
+      "epoch": 6.895424836601308,
+      "grad_norm": 1.5469038486480713,
+      "learning_rate": 0.0002,
+      "loss": 0.9177,
+      "step": 10550
+    },
+    {
+      "epoch": 6.901960784313726,
+      "grad_norm": 1.1731038093566895,
+      "learning_rate": 0.0002,
+      "loss": 0.8545,
+      "step": 10560
+    },
+    {
+      "epoch": 6.908496732026144,
+      "grad_norm": 0.968008816242218,
+      "learning_rate": 0.0002,
+      "loss": 0.9795,
+      "step": 10570
+    },
+    {
+      "epoch": 6.915032679738562,
+      "grad_norm": 0.9082416892051697,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 10580
+    },
+    {
+      "epoch": 6.921568627450981,
+      "grad_norm": 1.5816899538040161,
+      "learning_rate": 0.0002,
+      "loss": 0.9898,
+      "step": 10590
+    },
+    {
+      "epoch": 6.928104575163399,
+      "grad_norm": 0.9462234377861023,
+      "learning_rate": 0.0002,
+      "loss": 0.9692,
+      "step": 10600
+    },
+    {
+      "epoch": 6.934640522875817,
+      "grad_norm": 1.4950200319290161,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 10610
+    },
+    {
+      "epoch": 6.9411764705882355,
+      "grad_norm": 1.2929182052612305,
+      "learning_rate": 0.0002,
+      "loss": 0.8888,
+      "step": 10620
+    },
+    {
+      "epoch": 6.947712418300654,
+      "grad_norm": 1.2995754480361938,
+      "learning_rate": 0.0002,
+      "loss": 1.0141,
+      "step": 10630
+    },
+    {
+      "epoch": 6.954248366013072,
+      "grad_norm": 0.9407122135162354,
+      "learning_rate": 0.0002,
+      "loss": 0.9863,
+      "step": 10640
+    },
+    {
+      "epoch": 6.96078431372549,
+      "grad_norm": 1.1735378503799438,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 10650
+    },
+    {
+      "epoch": 6.967320261437909,
+      "grad_norm": 0.9937344193458557,
+      "learning_rate": 0.0002,
+      "loss": 0.936,
+      "step": 10660
+    },
+    {
+      "epoch": 6.973856209150327,
+      "grad_norm": 1.2498728036880493,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 10670
+    },
+    {
+      "epoch": 6.980392156862745,
+      "grad_norm": 1.0513341426849365,
+      "learning_rate": 0.0002,
+      "loss": 1.0504,
+      "step": 10680
+    },
+    {
+      "epoch": 6.9869281045751634,
+      "grad_norm": 1.4611467123031616,
+      "learning_rate": 0.0002,
+      "loss": 0.9259,
+      "step": 10690
+    },
+    {
+      "epoch": 6.993464052287582,
+      "grad_norm": 1.2924799919128418,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 10700
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.2024929523468018,
+      "learning_rate": 0.0002,
+      "loss": 0.8953,
+      "step": 10710
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.4972445964813232,
+      "eval_runtime": 33.6225,
+      "eval_samples_per_second": 12.967,
+      "eval_steps_per_second": 1.636,
+      "step": 10710
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 1.3089758074896384e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-10710/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2f3483670b3b00ac0ed25619f0eb0c0b7112637a
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6b24de5485ae306b5ee4c7c4a1602e656d518d50ee3702da6aaec99610e4485
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e54e9f92fac05bb989d1f470ae4f0352aef0a4b3
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bde8adc9844a4c4ade5d56781e28a8b72ff6c6ae4f98411f82fe68cdd6a6e416
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ce4b2ba7fe1da609c7bfdda3f3ff828aa4fb8618
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01f5fa33db12a535a6a8d2cad85866729a0401e18eeecab9c6ee76c55d5da392
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..093465e351511c3a482f0ece5eb88f5b5d01d841
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f3f90db975d017768edcd4c27b014f37a6b88072ed40497cd25be5a074c7da
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8d96ba531cd696f91804f09d3c1d441a9a723cd8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/trainer_state.json
@@ -0,0 +1,8665 @@
+{
+  "best_metric": 1.4113320112228394,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120",
+  "epoch": 8.0,
+  "eval_steps": 10,
+  "global_step": 12240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0065359477124183,
+      "grad_norm": 0.6926420331001282,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0130718954248366,
+      "grad_norm": 0.6656355857849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 4610
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 0.9901936650276184,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 4620
+    },
+    {
+      "epoch": 3.026143790849673,
+      "grad_norm": 0.6713474988937378,
+      "learning_rate": 0.0002,
+      "loss": 1.22,
+      "step": 4630
+    },
+    {
+      "epoch": 3.0326797385620914,
+      "grad_norm": 0.6199324131011963,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 4640
+    },
+    {
+      "epoch": 3.0392156862745097,
+      "grad_norm": 0.7180785536766052,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 4650
+    },
+    {
+      "epoch": 3.045751633986928,
+      "grad_norm": 0.8256588578224182,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 4660
+    },
+    {
+      "epoch": 3.052287581699346,
+      "grad_norm": 0.6637389063835144,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 4670
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 0.6980698108673096,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 4680
+    },
+    {
+      "epoch": 3.065359477124183,
+      "grad_norm": 0.8091534972190857,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 4690
+    },
+    {
+      "epoch": 3.0718954248366015,
+      "grad_norm": 0.5715174078941345,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 4700
+    },
+    {
+      "epoch": 3.0784313725490198,
+      "grad_norm": 0.735639750957489,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 4710
+    },
+    {
+      "epoch": 3.084967320261438,
+      "grad_norm": 0.7619708180427551,
+      "learning_rate": 0.0002,
+      "loss": 1.1522,
+      "step": 4720
+    },
+    {
+      "epoch": 3.0915032679738563,
+      "grad_norm": 1.263566017150879,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 4730
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 0.6600871682167053,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4740
+    },
+    {
+      "epoch": 3.104575163398693,
+      "grad_norm": 0.717792809009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 4750
+    },
+    {
+      "epoch": 3.111111111111111,
+      "grad_norm": 0.853714644908905,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1176470588235294,
+      "grad_norm": 1.1004153490066528,
+      "learning_rate": 0.0002,
+      "loss": 1.2031,
+      "step": 4770
+    },
+    {
+      "epoch": 3.1241830065359477,
+      "grad_norm": 0.8566235899925232,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 4780
+    },
+    {
+      "epoch": 3.130718954248366,
+      "grad_norm": 0.8315296173095703,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 4790
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 0.8020524978637695,
+      "learning_rate": 0.0002,
+      "loss": 1.32,
+      "step": 4800
+    },
+    {
+      "epoch": 3.1437908496732025,
+      "grad_norm": 0.7564275860786438,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 4810
+    },
+    {
+      "epoch": 3.150326797385621,
+      "grad_norm": 0.9077776670455933,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 4820
+    },
+    {
+      "epoch": 3.156862745098039,
+      "grad_norm": 0.6323099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 4830
+    },
+    {
+      "epoch": 3.1633986928104574,
+      "grad_norm": 0.6625368595123291,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 4840
+    },
+    {
+      "epoch": 3.1699346405228757,
+      "grad_norm": 0.8119261860847473,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 4850
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 0.6399450898170471,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 4860
+    },
+    {
+      "epoch": 3.183006535947712,
+      "grad_norm": 1.0659016370773315,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 4870
+    },
+    {
+      "epoch": 3.189542483660131,
+      "grad_norm": 0.8040369749069214,
+      "learning_rate": 0.0002,
+      "loss": 1.2914,
+      "step": 4880
+    },
+    {
+      "epoch": 3.196078431372549,
+      "grad_norm": 0.7784733176231384,
+      "learning_rate": 0.0002,
+      "loss": 1.1996,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2026143790849675,
+      "grad_norm": 0.9660294651985168,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 4900
+    },
+    {
+      "epoch": 3.2091503267973858,
+      "grad_norm": 1.0676977634429932,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 4910
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 0.5877565741539001,
+      "learning_rate": 0.0002,
+      "loss": 1.0083,
+      "step": 4920
+    },
+    {
+      "epoch": 3.2222222222222223,
+      "grad_norm": 0.6164032816886902,
+      "learning_rate": 0.0002,
+      "loss": 1.1046,
+      "step": 4930
+    },
+    {
+      "epoch": 3.2287581699346406,
+      "grad_norm": 0.7627606987953186,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 4940
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.7442803978919983,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 4950
+    },
+    {
+      "epoch": 3.241830065359477,
+      "grad_norm": 0.7277812361717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 4960
+    },
+    {
+      "epoch": 3.2483660130718954,
+      "grad_norm": 1.0301902294158936,
+      "learning_rate": 0.0002,
+      "loss": 1.2237,
+      "step": 4970
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 0.7798232436180115,
+      "learning_rate": 0.0002,
+      "loss": 1.1466,
+      "step": 4980
+    },
+    {
+      "epoch": 3.261437908496732,
+      "grad_norm": 1.210265874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.2142,
+      "step": 4990
+    },
+    {
+      "epoch": 3.2679738562091503,
+      "grad_norm": 0.6677713990211487,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2745098039215685,
+      "grad_norm": 1.0524500608444214,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 5010
+    },
+    {
+      "epoch": 3.281045751633987,
+      "grad_norm": 0.7091745734214783,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5020
+    },
+    {
+      "epoch": 3.287581699346405,
+      "grad_norm": 0.8523224592208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1891,
+      "step": 5030
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 0.6120608448982239,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3006535947712417,
+      "grad_norm": 0.7437472939491272,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 5050
+    },
+    {
+      "epoch": 3.30718954248366,
+      "grad_norm": 0.7611715197563171,
+      "learning_rate": 0.0002,
+      "loss": 1.1295,
+      "step": 5060
+    },
+    {
+      "epoch": 3.313725490196078,
+      "grad_norm": 0.7249704003334045,
+      "learning_rate": 0.0002,
+      "loss": 1.0531,
+      "step": 5070
+    },
+    {
+      "epoch": 3.3202614379084965,
+      "grad_norm": 0.7316247820854187,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5080
+    },
+    {
+      "epoch": 3.326797385620915,
+      "grad_norm": 0.562412440776825,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5090
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.7052176594734192,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5100
+    },
+    {
+      "epoch": 3.3398692810457518,
+      "grad_norm": 0.7714211344718933,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 5110
+    },
+    {
+      "epoch": 3.34640522875817,
+      "grad_norm": 1.0436055660247803,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5120
+    },
+    {
+      "epoch": 3.3529411764705883,
+      "grad_norm": 0.8867271542549133,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 5130
+    },
+    {
+      "epoch": 3.3594771241830066,
+      "grad_norm": 0.8371267914772034,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 5140
+    },
+    {
+      "epoch": 3.366013071895425,
+      "grad_norm": 0.7257837057113647,
+      "learning_rate": 0.0002,
+      "loss": 1.1073,
+      "step": 5150
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 0.7102002501487732,
+      "learning_rate": 0.0002,
+      "loss": 1.1162,
+      "step": 5160
+    },
+    {
+      "epoch": 3.3790849673202614,
+      "grad_norm": 0.7636350393295288,
+      "learning_rate": 0.0002,
+      "loss": 1.2056,
+      "step": 5170
+    },
+    {
+      "epoch": 3.3856209150326797,
+      "grad_norm": 0.6887359619140625,
+      "learning_rate": 0.0002,
+      "loss": 1.0708,
+      "step": 5180
+    },
+    {
+      "epoch": 3.392156862745098,
+      "grad_norm": 0.8141424655914307,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 5190
+    },
+    {
+      "epoch": 3.3986928104575163,
+      "grad_norm": 0.694423496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5200
+    },
+    {
+      "epoch": 3.4052287581699345,
+      "grad_norm": 0.914013683795929,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5210
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 0.8503239750862122,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 5220
+    },
+    {
+      "epoch": 3.418300653594771,
+      "grad_norm": 0.6196836233139038,
+      "learning_rate": 0.0002,
+      "loss": 1.2696,
+      "step": 5230
+    },
+    {
+      "epoch": 3.4248366013071894,
+      "grad_norm": 1.0760811567306519,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 5240
+    },
+    {
+      "epoch": 3.431372549019608,
+      "grad_norm": 0.6524698138237,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 5250
+    },
+    {
+      "epoch": 3.4379084967320264,
+      "grad_norm": 0.674467921257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5260
+    },
+    {
+      "epoch": 3.4444444444444446,
+      "grad_norm": 0.7690372467041016,
+      "learning_rate": 0.0002,
+      "loss": 1.1015,
+      "step": 5270
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 0.8751813769340515,
+      "learning_rate": 0.0002,
+      "loss": 1.2511,
+      "step": 5280
+    },
+    {
+      "epoch": 3.457516339869281,
+      "grad_norm": 0.750407874584198,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 5290
+    },
+    {
+      "epoch": 3.4640522875816995,
+      "grad_norm": 0.5991823077201843,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 5300
+    },
+    {
+      "epoch": 3.4705882352941178,
+      "grad_norm": 1.0164772272109985,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 5310
+    },
+    {
+      "epoch": 3.477124183006536,
+      "grad_norm": 0.8704105019569397,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 5320
+    },
+    {
+      "epoch": 3.4836601307189543,
+      "grad_norm": 0.709102213382721,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 5330
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 0.6273632049560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 5340
+    },
+    {
+      "epoch": 3.496732026143791,
+      "grad_norm": 0.6807359457015991,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 5350
+    },
+    {
+      "epoch": 3.503267973856209,
+      "grad_norm": 0.7085188627243042,
+      "learning_rate": 0.0002,
+      "loss": 1.131,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5098039215686274,
+      "grad_norm": 0.6938307881355286,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 5370
+    },
+    {
+      "epoch": 3.5163398692810457,
+      "grad_norm": 0.8544146418571472,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 5380
+    },
+    {
+      "epoch": 3.522875816993464,
+      "grad_norm": 0.7889642119407654,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 5390
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.7858421206474304,
+      "learning_rate": 0.0002,
+      "loss": 1.1691,
+      "step": 5400
+    },
+    {
+      "epoch": 3.5359477124183005,
+      "grad_norm": 0.8547123074531555,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5410
+    },
+    {
+      "epoch": 3.542483660130719,
+      "grad_norm": 0.8218181133270264,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 5420
+    },
+    {
+      "epoch": 3.549019607843137,
+      "grad_norm": 1.153623342514038,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 5430
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.1321099996566772,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 5440
+    },
+    {
+      "epoch": 3.5620915032679736,
+      "grad_norm": 0.9495334029197693,
+      "learning_rate": 0.0002,
+      "loss": 1.2224,
+      "step": 5450
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 0.8743821978569031,
+      "learning_rate": 0.0002,
+      "loss": 1.2869,
+      "step": 5460
+    },
+    {
+      "epoch": 3.57516339869281,
+      "grad_norm": 0.7513086795806885,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 5470
+    },
+    {
+      "epoch": 3.581699346405229,
+      "grad_norm": 1.0139480829238892,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 5480
+    },
+    {
+      "epoch": 3.588235294117647,
+      "grad_norm": 0.6615135073661804,
+      "learning_rate": 0.0002,
+      "loss": 1.1706,
+      "step": 5490
+    },
+    {
+      "epoch": 3.5947712418300655,
+      "grad_norm": 1.180798888206482,
+      "learning_rate": 0.0002,
+      "loss": 1.3906,
+      "step": 5500
+    },
+    {
+      "epoch": 3.6013071895424837,
+      "grad_norm": 0.7085279226303101,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 5510
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 0.540268063545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1623,
+      "step": 5520
+    },
+    {
+      "epoch": 3.6143790849673203,
+      "grad_norm": 0.7905671000480652,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 5530
+    },
+    {
+      "epoch": 3.6209150326797386,
+      "grad_norm": 0.8457717299461365,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 5540
+    },
+    {
+      "epoch": 3.627450980392157,
+      "grad_norm": 0.7102677822113037,
+      "learning_rate": 0.0002,
+      "loss": 1.1799,
+      "step": 5550
+    },
+    {
+      "epoch": 3.633986928104575,
+      "grad_norm": 0.7179514765739441,
+      "learning_rate": 0.0002,
+      "loss": 1.2394,
+      "step": 5560
+    },
+    {
+      "epoch": 3.6405228758169934,
+      "grad_norm": 1.0854148864746094,
+      "learning_rate": 0.0002,
+      "loss": 1.2019,
+      "step": 5570
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 0.8209951519966125,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5580
+    },
+    {
+      "epoch": 3.65359477124183,
+      "grad_norm": 0.6944138407707214,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 5590
+    },
+    {
+      "epoch": 3.6601307189542482,
+      "grad_norm": 0.7675473093986511,
+      "learning_rate": 0.0002,
+      "loss": 1.3226,
+      "step": 5600
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6683364510536194,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 5610
+    },
+    {
+      "epoch": 3.6732026143790852,
+      "grad_norm": 0.7920727133750916,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 5620
+    },
+    {
+      "epoch": 3.6797385620915035,
+      "grad_norm": 0.9440218806266785,
+      "learning_rate": 0.0002,
+      "loss": 1.2287,
+      "step": 5630
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 0.6600824594497681,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 5640
+    },
+    {
+      "epoch": 3.69281045751634,
+      "grad_norm": 0.6860619187355042,
+      "learning_rate": 0.0002,
+      "loss": 1.191,
+      "step": 5650
+    },
+    {
+      "epoch": 3.6993464052287583,
+      "grad_norm": 0.6579713225364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 5660
+    },
+    {
+      "epoch": 3.7058823529411766,
+      "grad_norm": 0.661081075668335,
+      "learning_rate": 0.0002,
+      "loss": 1.1464,
+      "step": 5670
+    },
+    {
+      "epoch": 3.712418300653595,
+      "grad_norm": 1.0968825817108154,
+      "learning_rate": 0.0002,
+      "loss": 1.289,
+      "step": 5680
+    },
+    {
+      "epoch": 3.718954248366013,
+      "grad_norm": 0.8066844940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.192,
+      "step": 5690
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 0.8341682553291321,
+      "learning_rate": 0.0002,
+      "loss": 1.2322,
+      "step": 5700
+    },
+    {
+      "epoch": 3.7320261437908497,
+      "grad_norm": 0.6682852506637573,
+      "learning_rate": 0.0002,
+      "loss": 1.1473,
+      "step": 5710
+    },
+    {
+      "epoch": 3.738562091503268,
+      "grad_norm": 0.898595929145813,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 5720
+    },
+    {
+      "epoch": 3.7450980392156863,
+      "grad_norm": 0.6876054406166077,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 5730
+    },
+    {
+      "epoch": 3.7516339869281046,
+      "grad_norm": 0.7817103266716003,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 5740
+    },
+    {
+      "epoch": 3.758169934640523,
+      "grad_norm": 0.5840168595314026,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 5750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 0.6263918876647949,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5760
+    },
+    {
+      "epoch": 3.7712418300653594,
+      "grad_norm": 0.7948952317237854,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 5770
+    },
+    {
+      "epoch": 3.7777777777777777,
+      "grad_norm": 0.6700998544692993,
+      "learning_rate": 0.0002,
+      "loss": 1.149,
+      "step": 5780
+    },
+    {
+      "epoch": 3.784313725490196,
+      "grad_norm": 1.1169519424438477,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 5790
+    },
+    {
+      "epoch": 3.7908496732026142,
+      "grad_norm": 0.8354471325874329,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 5800
+    },
+    {
+      "epoch": 3.7973856209150325,
+      "grad_norm": 0.6304181814193726,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 5810
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 0.6919655799865723,
+      "learning_rate": 0.0002,
+      "loss": 1.2059,
+      "step": 5820
+    },
+    {
+      "epoch": 3.810457516339869,
+      "grad_norm": 0.600385844707489,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5830
+    },
+    {
+      "epoch": 3.8169934640522873,
+      "grad_norm": 0.8406319618225098,
+      "learning_rate": 0.0002,
+      "loss": 1.2324,
+      "step": 5840
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.7594282031059265,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5850
+    },
+    {
+      "epoch": 3.8300653594771243,
+      "grad_norm": 0.8179879784584045,
+      "learning_rate": 0.0002,
+      "loss": 1.1903,
+      "step": 5860
+    },
+    {
+      "epoch": 3.8366013071895426,
+      "grad_norm": 1.141430377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 5870
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 0.6595550775527954,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 5880
+    },
+    {
+      "epoch": 3.849673202614379,
+      "grad_norm": 0.7499435544013977,
+      "learning_rate": 0.0002,
+      "loss": 1.2378,
+      "step": 5890
+    },
+    {
+      "epoch": 3.8562091503267975,
+      "grad_norm": 0.7851517200469971,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5900
+    },
+    {
+      "epoch": 3.8627450980392157,
+      "grad_norm": 1.0533545017242432,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 5910
+    },
+    {
+      "epoch": 3.869281045751634,
+      "grad_norm": 0.960086464881897,
+      "learning_rate": 0.0002,
+      "loss": 1.3576,
+      "step": 5920
+    },
+    {
+      "epoch": 3.8758169934640523,
+      "grad_norm": 0.9952049851417542,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 5930
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 0.7884191274642944,
+      "learning_rate": 0.0002,
+      "loss": 1.2027,
+      "step": 5940
+    },
+    {
+      "epoch": 3.888888888888889,
+      "grad_norm": 0.7461766600608826,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5950
+    },
+    {
+      "epoch": 3.895424836601307,
+      "grad_norm": 0.9594355821609497,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 5960
+    },
+    {
+      "epoch": 3.9019607843137254,
+      "grad_norm": 0.8179471492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.1164,
+      "step": 5970
+    },
+    {
+      "epoch": 3.9084967320261437,
+      "grad_norm": 0.8240267634391785,
+      "learning_rate": 0.0002,
+      "loss": 1.2421,
+      "step": 5980
+    },
+    {
+      "epoch": 3.915032679738562,
+      "grad_norm": 0.7462618350982666,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 5990
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 0.711207389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.2124,
+      "step": 6000
+    },
+    {
+      "epoch": 3.928104575163399,
+      "grad_norm": 0.6910956501960754,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 6010
+    },
+    {
+      "epoch": 3.9346405228758172,
+      "grad_norm": 0.749093770980835,
+      "learning_rate": 0.0002,
+      "loss": 1.2127,
+      "step": 6020
+    },
+    {
+      "epoch": 3.9411764705882355,
+      "grad_norm": 1.3332762718200684,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6030
+    },
+    {
+      "epoch": 3.947712418300654,
+      "grad_norm": 0.71457439661026,
+      "learning_rate": 0.0002,
+      "loss": 1.1442,
+      "step": 6040
+    },
+    {
+      "epoch": 3.954248366013072,
+      "grad_norm": 1.1205238103866577,
+      "learning_rate": 0.0002,
+      "loss": 1.339,
+      "step": 6050
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 0.6958928108215332,
+      "learning_rate": 0.0002,
+      "loss": 1.2962,
+      "step": 6060
+    },
+    {
+      "epoch": 3.9673202614379086,
+      "grad_norm": 0.7518056035041809,
+      "learning_rate": 0.0002,
+      "loss": 1.1802,
+      "step": 6070
+    },
+    {
+      "epoch": 3.973856209150327,
+      "grad_norm": 0.8010755777359009,
+      "learning_rate": 0.0002,
+      "loss": 1.1179,
+      "step": 6080
+    },
+    {
+      "epoch": 3.980392156862745,
+      "grad_norm": 0.7492658495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 6090
+    },
+    {
+      "epoch": 3.9869281045751634,
+      "grad_norm": 0.900704562664032,
+      "learning_rate": 0.0002,
+      "loss": 1.2113,
+      "step": 6100
+    },
+    {
+      "epoch": 3.9934640522875817,
+      "grad_norm": 0.7997331619262695,
+      "learning_rate": 0.0002,
+      "loss": 1.1106,
+      "step": 6110
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7163209319114685,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 6120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4113320112228394,
+      "eval_runtime": 33.7199,
+      "eval_samples_per_second": 12.93,
+      "eval_steps_per_second": 1.631,
+      "step": 6120
+    },
+    {
+      "epoch": 4.006535947712418,
+      "grad_norm": 0.9527022838592529,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 6130
+    },
+    {
+      "epoch": 4.0130718954248366,
+      "grad_norm": 0.7603210210800171,
+      "learning_rate": 0.0002,
+      "loss": 1.101,
+      "step": 6140
+    },
+    {
+      "epoch": 4.019607843137255,
+      "grad_norm": 1.127387523651123,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 6150
+    },
+    {
+      "epoch": 4.026143790849673,
+      "grad_norm": 0.8290133476257324,
+      "learning_rate": 0.0002,
+      "loss": 1.0734,
+      "step": 6160
+    },
+    {
+      "epoch": 4.032679738562091,
+      "grad_norm": 0.9912241101264954,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 6170
+    },
+    {
+      "epoch": 4.03921568627451,
+      "grad_norm": 0.947005033493042,
+      "learning_rate": 0.0002,
+      "loss": 1.0719,
+      "step": 6180
+    },
+    {
+      "epoch": 4.045751633986928,
+      "grad_norm": 0.707466185092926,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 6190
+    },
+    {
+      "epoch": 4.052287581699346,
+      "grad_norm": 1.0604327917099,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6200
+    },
+    {
+      "epoch": 4.0588235294117645,
+      "grad_norm": 0.7848685383796692,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6210
+    },
+    {
+      "epoch": 4.065359477124183,
+      "grad_norm": 0.8475256562232971,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 6220
+    },
+    {
+      "epoch": 4.071895424836601,
+      "grad_norm": 0.9759448766708374,
+      "learning_rate": 0.0002,
+      "loss": 1.1104,
+      "step": 6230
+    },
+    {
+      "epoch": 4.078431372549019,
+      "grad_norm": 0.9324519038200378,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 6240
+    },
+    {
+      "epoch": 4.084967320261438,
+      "grad_norm": 0.8723901510238647,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 6250
+    },
+    {
+      "epoch": 4.091503267973856,
+      "grad_norm": 0.8343415856361389,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6260
+    },
+    {
+      "epoch": 4.098039215686274,
+      "grad_norm": 0.7490310072898865,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 6270
+    },
+    {
+      "epoch": 4.104575163398692,
+      "grad_norm": 0.8961182832717896,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 6280
+    },
+    {
+      "epoch": 4.111111111111111,
+      "grad_norm": 0.7124854922294617,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 6290
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 0.8338138461112976,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 6300
+    },
+    {
+      "epoch": 4.124183006535947,
+      "grad_norm": 0.8075833320617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 6310
+    },
+    {
+      "epoch": 4.130718954248366,
+      "grad_norm": 0.8069391846656799,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 6320
+    },
+    {
+      "epoch": 4.137254901960785,
+      "grad_norm": 0.9567893147468567,
+      "learning_rate": 0.0002,
+      "loss": 0.948,
+      "step": 6330
+    },
+    {
+      "epoch": 4.143790849673203,
+      "grad_norm": 1.2184662818908691,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 6340
+    },
+    {
+      "epoch": 4.150326797385621,
+      "grad_norm": 1.030976414680481,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 6350
+    },
+    {
+      "epoch": 4.1568627450980395,
+      "grad_norm": 0.9749957323074341,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6360
+    },
+    {
+      "epoch": 4.163398692810458,
+      "grad_norm": 0.7089483141899109,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 6370
+    },
+    {
+      "epoch": 4.169934640522876,
+      "grad_norm": 1.1084946393966675,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 6380
+    },
+    {
+      "epoch": 4.176470588235294,
+      "grad_norm": 0.7998497486114502,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 6390
+    },
+    {
+      "epoch": 4.183006535947713,
+      "grad_norm": 0.8997811675071716,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 6400
+    },
+    {
+      "epoch": 4.189542483660131,
+      "grad_norm": 0.8359479904174805,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 6410
+    },
+    {
+      "epoch": 4.196078431372549,
+      "grad_norm": 0.9087472558021545,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 6420
+    },
+    {
+      "epoch": 4.2026143790849675,
+      "grad_norm": 1.1100451946258545,
+      "learning_rate": 0.0002,
+      "loss": 1.0657,
+      "step": 6430
+    },
+    {
+      "epoch": 4.209150326797386,
+      "grad_norm": 0.9376999735832214,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 6440
+    },
+    {
+      "epoch": 4.215686274509804,
+      "grad_norm": 0.8179266452789307,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 6450
+    },
+    {
+      "epoch": 4.222222222222222,
+      "grad_norm": 0.9953271746635437,
+      "learning_rate": 0.0002,
+      "loss": 1.0679,
+      "step": 6460
+    },
+    {
+      "epoch": 4.228758169934641,
+      "grad_norm": 0.8476650714874268,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 6470
+    },
+    {
+      "epoch": 4.235294117647059,
+      "grad_norm": 0.8406323194503784,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 6480
+    },
+    {
+      "epoch": 4.241830065359477,
+      "grad_norm": 0.819134533405304,
+      "learning_rate": 0.0002,
+      "loss": 1.057,
+      "step": 6490
+    },
+    {
+      "epoch": 4.248366013071895,
+      "grad_norm": 0.7764983773231506,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 6500
+    },
+    {
+      "epoch": 4.254901960784314,
+      "grad_norm": 0.8252112865447998,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 6510
+    },
+    {
+      "epoch": 4.261437908496732,
+      "grad_norm": 0.7941019535064697,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 6520
+    },
+    {
+      "epoch": 4.26797385620915,
+      "grad_norm": 0.7673905491828918,
+      "learning_rate": 0.0002,
+      "loss": 1.0296,
+      "step": 6530
+    },
+    {
+      "epoch": 4.2745098039215685,
+      "grad_norm": 0.8749890327453613,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6540
+    },
+    {
+      "epoch": 4.281045751633987,
+      "grad_norm": 0.7343207597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 6550
+    },
+    {
+      "epoch": 4.287581699346405,
+      "grad_norm": 1.2786651849746704,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6560
+    },
+    {
+      "epoch": 4.294117647058823,
+      "grad_norm": 1.316875696182251,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6570
+    },
+    {
+      "epoch": 4.300653594771242,
+      "grad_norm": 0.8349189162254333,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 6580
+    },
+    {
+      "epoch": 4.30718954248366,
+      "grad_norm": 0.7510647177696228,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6590
+    },
+    {
+      "epoch": 4.313725490196078,
+      "grad_norm": 0.932420551776886,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6600
+    },
+    {
+      "epoch": 4.3202614379084965,
+      "grad_norm": 0.8510616421699524,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 6610
+    },
+    {
+      "epoch": 4.326797385620915,
+      "grad_norm": 0.7661547064781189,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6620
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 1.0370930433273315,
+      "learning_rate": 0.0002,
+      "loss": 1.2064,
+      "step": 6630
+    },
+    {
+      "epoch": 4.339869281045751,
+      "grad_norm": 0.9302158951759338,
+      "learning_rate": 0.0002,
+      "loss": 1.1064,
+      "step": 6640
+    },
+    {
+      "epoch": 4.34640522875817,
+      "grad_norm": 0.9203811883926392,
+      "learning_rate": 0.0002,
+      "loss": 0.968,
+      "step": 6650
+    },
+    {
+      "epoch": 4.352941176470588,
+      "grad_norm": 0.9986332654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 6660
+    },
+    {
+      "epoch": 4.359477124183006,
+      "grad_norm": 0.8001713156700134,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6670
+    },
+    {
+      "epoch": 4.366013071895424,
+      "grad_norm": 0.829714298248291,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 6680
+    },
+    {
+      "epoch": 4.372549019607844,
+      "grad_norm": 0.8253079056739807,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 6690
+    },
+    {
+      "epoch": 4.379084967320262,
+      "grad_norm": 0.824666440486908,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 6700
+    },
+    {
+      "epoch": 4.38562091503268,
+      "grad_norm": 0.8872972130775452,
+      "learning_rate": 0.0002,
+      "loss": 1.1968,
+      "step": 6710
+    },
+    {
+      "epoch": 4.392156862745098,
+      "grad_norm": 0.8729761838912964,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 6720
+    },
+    {
+      "epoch": 4.398692810457517,
+      "grad_norm": 1.1367264986038208,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 6730
+    },
+    {
+      "epoch": 4.405228758169935,
+      "grad_norm": 0.9699058532714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0184,
+      "step": 6740
+    },
+    {
+      "epoch": 4.411764705882353,
+      "grad_norm": 0.8266763687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 6750
+    },
+    {
+      "epoch": 4.4183006535947715,
+      "grad_norm": 1.0249767303466797,
+      "learning_rate": 0.0002,
+      "loss": 1.0735,
+      "step": 6760
+    },
+    {
+      "epoch": 4.42483660130719,
+      "grad_norm": 0.73606938123703,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 6770
+    },
+    {
+      "epoch": 4.431372549019608,
+      "grad_norm": 1.4050679206848145,
+      "learning_rate": 0.0002,
+      "loss": 1.1037,
+      "step": 6780
+    },
+    {
+      "epoch": 4.437908496732026,
+      "grad_norm": 1.1114081144332886,
+      "learning_rate": 0.0002,
+      "loss": 1.1418,
+      "step": 6790
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8031067848205566,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 6800
+    },
+    {
+      "epoch": 4.450980392156863,
+      "grad_norm": 0.8513566851615906,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6810
+    },
+    {
+      "epoch": 4.457516339869281,
+      "grad_norm": 1.332741379737854,
+      "learning_rate": 0.0002,
+      "loss": 1.1852,
+      "step": 6820
+    },
+    {
+      "epoch": 4.4640522875816995,
+      "grad_norm": 1.5032578706741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 6830
+    },
+    {
+      "epoch": 4.470588235294118,
+      "grad_norm": 0.7677283883094788,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6840
+    },
+    {
+      "epoch": 4.477124183006536,
+      "grad_norm": 0.989148736000061,
+      "learning_rate": 0.0002,
+      "loss": 1.1501,
+      "step": 6850
+    },
+    {
+      "epoch": 4.483660130718954,
+      "grad_norm": 1.5316275358200073,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 6860
+    },
+    {
+      "epoch": 4.490196078431373,
+      "grad_norm": 0.9427124261856079,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 6870
+    },
+    {
+      "epoch": 4.496732026143791,
+      "grad_norm": 1.215287685394287,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 6880
+    },
+    {
+      "epoch": 4.503267973856209,
+      "grad_norm": 0.7286760210990906,
+      "learning_rate": 0.0002,
+      "loss": 1.0809,
+      "step": 6890
+    },
+    {
+      "epoch": 4.509803921568627,
+      "grad_norm": 0.874829888343811,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6900
+    },
+    {
+      "epoch": 4.516339869281046,
+      "grad_norm": 0.8058359622955322,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 6910
+    },
+    {
+      "epoch": 4.522875816993464,
+      "grad_norm": 1.248195767402649,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 6920
+    },
+    {
+      "epoch": 4.529411764705882,
+      "grad_norm": 0.8033645749092102,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 6930
+    },
+    {
+      "epoch": 4.5359477124183005,
+      "grad_norm": 1.7361950874328613,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 6940
+    },
+    {
+      "epoch": 4.542483660130719,
+      "grad_norm": 0.8058095574378967,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 6950
+    },
+    {
+      "epoch": 4.549019607843137,
+      "grad_norm": 1.254089593887329,
+      "learning_rate": 0.0002,
+      "loss": 1.0057,
+      "step": 6960
+    },
+    {
+      "epoch": 4.555555555555555,
+      "grad_norm": 0.9180455803871155,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 6970
+    },
+    {
+      "epoch": 4.562091503267974,
+      "grad_norm": 0.6677682399749756,
+      "learning_rate": 0.0002,
+      "loss": 1.0559,
+      "step": 6980
+    },
+    {
+      "epoch": 4.568627450980392,
+      "grad_norm": 0.8127354383468628,
+      "learning_rate": 0.0002,
+      "loss": 1.0453,
+      "step": 6990
+    },
+    {
+      "epoch": 4.57516339869281,
+      "grad_norm": 1.0263001918792725,
+      "learning_rate": 0.0002,
+      "loss": 1.0828,
+      "step": 7000
+    },
+    {
+      "epoch": 4.5816993464052285,
+      "grad_norm": 0.9641909003257751,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 7010
+    },
+    {
+      "epoch": 4.588235294117647,
+      "grad_norm": 0.9440861344337463,
+      "learning_rate": 0.0002,
+      "loss": 1.179,
+      "step": 7020
+    },
+    {
+      "epoch": 4.594771241830065,
+      "grad_norm": 0.9539011716842651,
+      "learning_rate": 0.0002,
+      "loss": 1.0931,
+      "step": 7030
+    },
+    {
+      "epoch": 4.601307189542483,
+      "grad_norm": 1.0449910163879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7040
+    },
+    {
+      "epoch": 4.607843137254902,
+      "grad_norm": 0.8766893744468689,
+      "learning_rate": 0.0002,
+      "loss": 0.9944,
+      "step": 7050
+    },
+    {
+      "epoch": 4.61437908496732,
+      "grad_norm": 0.6983462572097778,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 7060
+    },
+    {
+      "epoch": 4.620915032679738,
+      "grad_norm": 0.9505505561828613,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 7070
+    },
+    {
+      "epoch": 4.627450980392156,
+      "grad_norm": 1.2506657838821411,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7080
+    },
+    {
+      "epoch": 4.633986928104575,
+      "grad_norm": 0.9602801203727722,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 7090
+    },
+    {
+      "epoch": 4.640522875816993,
+      "grad_norm": 0.7398977875709534,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 7100
+    },
+    {
+      "epoch": 4.647058823529412,
+      "grad_norm": 1.3862425088882446,
+      "learning_rate": 0.0002,
+      "loss": 1.0769,
+      "step": 7110
+    },
+    {
+      "epoch": 4.65359477124183,
+      "grad_norm": 1.1451990604400635,
+      "learning_rate": 0.0002,
+      "loss": 1.0571,
+      "step": 7120
+    },
+    {
+      "epoch": 4.660130718954249,
+      "grad_norm": 0.9010422229766846,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 7130
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.7102518081665039,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 7140
+    },
+    {
+      "epoch": 4.673202614379085,
+      "grad_norm": 0.7963796257972717,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7150
+    },
+    {
+      "epoch": 4.6797385620915035,
+      "grad_norm": 0.7726007699966431,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 7160
+    },
+    {
+      "epoch": 4.686274509803922,
+      "grad_norm": 0.8097564578056335,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 7170
+    },
+    {
+      "epoch": 4.69281045751634,
+      "grad_norm": 0.9070925116539001,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 7180
+    },
+    {
+      "epoch": 4.699346405228758,
+      "grad_norm": 0.7543528079986572,
+      "learning_rate": 0.0002,
+      "loss": 1.0315,
+      "step": 7190
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.9900904893875122,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 7200
+    },
+    {
+      "epoch": 4.712418300653595,
+      "grad_norm": 0.8033412098884583,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 7210
+    },
+    {
+      "epoch": 4.718954248366013,
+      "grad_norm": 0.8440839052200317,
+      "learning_rate": 0.0002,
+      "loss": 1.1773,
+      "step": 7220
+    },
+    {
+      "epoch": 4.7254901960784315,
+      "grad_norm": 0.9325555562973022,
+      "learning_rate": 0.0002,
+      "loss": 1.1258,
+      "step": 7230
+    },
+    {
+      "epoch": 4.73202614379085,
+      "grad_norm": 0.7881146669387817,
+      "learning_rate": 0.0002,
+      "loss": 1.1384,
+      "step": 7240
+    },
+    {
+      "epoch": 4.738562091503268,
+      "grad_norm": 0.884453296661377,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7250
+    },
+    {
+      "epoch": 4.745098039215686,
+      "grad_norm": 0.9274539351463318,
+      "learning_rate": 0.0002,
+      "loss": 1.1036,
+      "step": 7260
+    },
+    {
+      "epoch": 4.751633986928105,
+      "grad_norm": 1.2367479801177979,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7270
+    },
+    {
+      "epoch": 4.758169934640523,
+      "grad_norm": 0.9499821066856384,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7280
+    },
+    {
+      "epoch": 4.764705882352941,
+      "grad_norm": 2.1918580532073975,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 7290
+    },
+    {
+      "epoch": 4.771241830065359,
+      "grad_norm": 0.8221880793571472,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 7300
+    },
+    {
+      "epoch": 4.777777777777778,
+      "grad_norm": 0.871972918510437,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7310
+    },
+    {
+      "epoch": 4.784313725490196,
+      "grad_norm": 0.8034510612487793,
+      "learning_rate": 0.0002,
+      "loss": 1.0599,
+      "step": 7320
+    },
+    {
+      "epoch": 4.790849673202614,
+      "grad_norm": 0.8959605693817139,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 7330
+    },
+    {
+      "epoch": 4.7973856209150325,
+      "grad_norm": 1.2326215505599976,
+      "learning_rate": 0.0002,
+      "loss": 1.0176,
+      "step": 7340
+    },
+    {
+      "epoch": 4.803921568627451,
+      "grad_norm": 0.9725791811943054,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7350
+    },
+    {
+      "epoch": 4.810457516339869,
+      "grad_norm": 0.7240816354751587,
+      "learning_rate": 0.0002,
+      "loss": 1.1229,
+      "step": 7360
+    },
+    {
+      "epoch": 4.816993464052287,
+      "grad_norm": 0.8265769481658936,
+      "learning_rate": 0.0002,
+      "loss": 1.0669,
+      "step": 7370
+    },
+    {
+      "epoch": 4.823529411764706,
+      "grad_norm": 0.8888696432113647,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 7380
+    },
+    {
+      "epoch": 4.830065359477124,
+      "grad_norm": 0.7776556015014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 7390
+    },
+    {
+      "epoch": 4.836601307189542,
+      "grad_norm": 0.8772371411323547,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7400
+    },
+    {
+      "epoch": 4.8431372549019605,
+      "grad_norm": 0.9786531925201416,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7410
+    },
+    {
+      "epoch": 4.849673202614379,
+      "grad_norm": 0.9059745073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7420
+    },
+    {
+      "epoch": 4.856209150326797,
+      "grad_norm": 0.7422552108764648,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 7430
+    },
+    {
+      "epoch": 4.862745098039216,
+      "grad_norm": 1.3040380477905273,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 7440
+    },
+    {
+      "epoch": 4.8692810457516345,
+      "grad_norm": 1.3278473615646362,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 7450
+    },
+    {
+      "epoch": 4.875816993464053,
+      "grad_norm": 1.2705849409103394,
+      "learning_rate": 0.0002,
+      "loss": 1.0713,
+      "step": 7460
+    },
+    {
+      "epoch": 4.882352941176471,
+      "grad_norm": 0.8837892413139343,
+      "learning_rate": 0.0002,
+      "loss": 1.0034,
+      "step": 7470
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 0.8670691251754761,
+      "learning_rate": 0.0002,
+      "loss": 1.1716,
+      "step": 7480
+    },
+    {
+      "epoch": 4.895424836601308,
+      "grad_norm": 0.9662758111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 7490
+    },
+    {
+      "epoch": 4.901960784313726,
+      "grad_norm": 0.8188302516937256,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 7500
+    },
+    {
+      "epoch": 4.908496732026144,
+      "grad_norm": 0.769442617893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 7510
+    },
+    {
+      "epoch": 4.915032679738562,
+      "grad_norm": 1.1465084552764893,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 7520
+    },
+    {
+      "epoch": 4.921568627450981,
+      "grad_norm": 1.253214955329895,
+      "learning_rate": 0.0002,
+      "loss": 1.0768,
+      "step": 7530
+    },
+    {
+      "epoch": 4.928104575163399,
+      "grad_norm": 0.7922375202178955,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 7540
+    },
+    {
+      "epoch": 4.934640522875817,
+      "grad_norm": 0.8306851387023926,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7550
+    },
+    {
+      "epoch": 4.9411764705882355,
+      "grad_norm": 0.8486151099205017,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 7560
+    },
+    {
+      "epoch": 4.947712418300654,
+      "grad_norm": 1.2601467370986938,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7570
+    },
+    {
+      "epoch": 4.954248366013072,
+      "grad_norm": 0.7980747818946838,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 7580
+    },
+    {
+      "epoch": 4.96078431372549,
+      "grad_norm": 0.8653254508972168,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 7590
+    },
+    {
+      "epoch": 4.967320261437909,
+      "grad_norm": 0.9680571556091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 7600
+    },
+    {
+      "epoch": 4.973856209150327,
+      "grad_norm": 0.9554466605186462,
+      "learning_rate": 0.0002,
+      "loss": 1.1795,
+      "step": 7610
+    },
+    {
+      "epoch": 4.980392156862745,
+      "grad_norm": 1.3693897724151611,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 7620
+    },
+    {
+      "epoch": 4.9869281045751634,
+      "grad_norm": 0.7809282541275024,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 7630
+    },
+    {
+      "epoch": 4.993464052287582,
+      "grad_norm": 0.7528006434440613,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 7640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7491309642791748,
+      "learning_rate": 0.0002,
+      "loss": 0.9951,
+      "step": 7650
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.4197258949279785,
+      "eval_runtime": 33.6327,
+      "eval_samples_per_second": 12.964,
+      "eval_steps_per_second": 1.635,
+      "step": 7650
+    },
+    {
+      "epoch": 5.006535947712418,
+      "grad_norm": 0.8840063214302063,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 7660
+    },
+    {
+      "epoch": 5.0130718954248366,
+      "grad_norm": 1.0118401050567627,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 7670
+    },
+    {
+      "epoch": 5.019607843137255,
+      "grad_norm": 1.0040518045425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1667,
+      "step": 7680
+    },
+    {
+      "epoch": 5.026143790849673,
+      "grad_norm": 0.7541199922561646,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 7690
+    },
+    {
+      "epoch": 5.032679738562091,
+      "grad_norm": 0.9106482863426208,
+      "learning_rate": 0.0002,
+      "loss": 1.0797,
+      "step": 7700
+    },
+    {
+      "epoch": 5.03921568627451,
+      "grad_norm": 1.3691469430923462,
+      "learning_rate": 0.0002,
+      "loss": 1.0096,
+      "step": 7710
+    },
+    {
+      "epoch": 5.045751633986928,
+      "grad_norm": 0.9449689388275146,
+      "learning_rate": 0.0002,
+      "loss": 0.9889,
+      "step": 7720
+    },
+    {
+      "epoch": 5.052287581699346,
+      "grad_norm": 1.1678508520126343,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 7730
+    },
+    {
+      "epoch": 5.0588235294117645,
+      "grad_norm": 1.1296145915985107,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7740
+    },
+    {
+      "epoch": 5.065359477124183,
+      "grad_norm": 0.7863904237747192,
+      "learning_rate": 0.0002,
+      "loss": 0.9339,
+      "step": 7750
+    },
+    {
+      "epoch": 5.071895424836601,
+      "grad_norm": 0.8691433072090149,
+      "learning_rate": 0.0002,
+      "loss": 1.0135,
+      "step": 7760
+    },
+    {
+      "epoch": 5.078431372549019,
+      "grad_norm": 1.0722088813781738,
+      "learning_rate": 0.0002,
+      "loss": 0.9776,
+      "step": 7770
+    },
+    {
+      "epoch": 5.084967320261438,
+      "grad_norm": 0.9625038504600525,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 7780
+    },
+    {
+      "epoch": 5.091503267973856,
+      "grad_norm": 1.2618783712387085,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 7790
+    },
+    {
+      "epoch": 5.098039215686274,
+      "grad_norm": 0.9970650672912598,
+      "learning_rate": 0.0002,
+      "loss": 0.9396,
+      "step": 7800
+    },
+    {
+      "epoch": 5.104575163398692,
+      "grad_norm": 1.3946677446365356,
+      "learning_rate": 0.0002,
+      "loss": 0.9186,
+      "step": 7810
+    },
+    {
+      "epoch": 5.111111111111111,
+      "grad_norm": 1.0260052680969238,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 7820
+    },
+    {
+      "epoch": 5.117647058823529,
+      "grad_norm": 1.105521559715271,
+      "learning_rate": 0.0002,
+      "loss": 0.9865,
+      "step": 7830
+    },
+    {
+      "epoch": 5.124183006535947,
+      "grad_norm": 1.003641128540039,
+      "learning_rate": 0.0002,
+      "loss": 0.9788,
+      "step": 7840
+    },
+    {
+      "epoch": 5.130718954248366,
+      "grad_norm": 1.0315021276474,
+      "learning_rate": 0.0002,
+      "loss": 0.9688,
+      "step": 7850
+    },
+    {
+      "epoch": 5.137254901960785,
+      "grad_norm": 0.9469530582427979,
+      "learning_rate": 0.0002,
+      "loss": 1.0001,
+      "step": 7860
+    },
+    {
+      "epoch": 5.143790849673203,
+      "grad_norm": 1.3244667053222656,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7870
+    },
+    {
+      "epoch": 5.150326797385621,
+      "grad_norm": 1.1732033491134644,
+      "learning_rate": 0.0002,
+      "loss": 0.9657,
+      "step": 7880
+    },
+    {
+      "epoch": 5.1568627450980395,
+      "grad_norm": 1.3129149675369263,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 7890
+    },
+    {
+      "epoch": 5.163398692810458,
+      "grad_norm": 0.8589454293251038,
+      "learning_rate": 0.0002,
+      "loss": 0.9894,
+      "step": 7900
+    },
+    {
+      "epoch": 5.169934640522876,
+      "grad_norm": 0.8954233527183533,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7910
+    },
+    {
+      "epoch": 5.176470588235294,
+      "grad_norm": 0.7426522970199585,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 7920
+    },
+    {
+      "epoch": 5.183006535947713,
+      "grad_norm": 1.1990121603012085,
+      "learning_rate": 0.0002,
+      "loss": 1.0106,
+      "step": 7930
+    },
+    {
+      "epoch": 5.189542483660131,
+      "grad_norm": 0.8867580890655518,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 7940
+    },
+    {
+      "epoch": 5.196078431372549,
+      "grad_norm": 1.016276478767395,
+      "learning_rate": 0.0002,
+      "loss": 0.9727,
+      "step": 7950
+    },
+    {
+      "epoch": 5.2026143790849675,
+      "grad_norm": 1.0210685729980469,
+      "learning_rate": 0.0002,
+      "loss": 0.9908,
+      "step": 7960
+    },
+    {
+      "epoch": 5.209150326797386,
+      "grad_norm": 1.0093122720718384,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7970
+    },
+    {
+      "epoch": 5.215686274509804,
+      "grad_norm": 0.9746801853179932,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 7980
+    },
+    {
+      "epoch": 5.222222222222222,
+      "grad_norm": 0.9113537073135376,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 7990
+    },
+    {
+      "epoch": 5.228758169934641,
+      "grad_norm": 1.2782206535339355,
+      "learning_rate": 0.0002,
+      "loss": 0.9167,
+      "step": 8000
+    },
+    {
+      "epoch": 5.235294117647059,
+      "grad_norm": 1.3223118782043457,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 8010
+    },
+    {
+      "epoch": 5.241830065359477,
+      "grad_norm": 0.7898629307746887,
+      "learning_rate": 0.0002,
+      "loss": 0.9244,
+      "step": 8020
+    },
+    {
+      "epoch": 5.248366013071895,
+      "grad_norm": 0.9822350740432739,
+      "learning_rate": 0.0002,
+      "loss": 1.0574,
+      "step": 8030
+    },
+    {
+      "epoch": 5.254901960784314,
+      "grad_norm": 1.5114340782165527,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 8040
+    },
+    {
+      "epoch": 5.261437908496732,
+      "grad_norm": 0.859006941318512,
+      "learning_rate": 0.0002,
+      "loss": 0.9816,
+      "step": 8050
+    },
+    {
+      "epoch": 5.26797385620915,
+      "grad_norm": 1.0495043992996216,
+      "learning_rate": 0.0002,
+      "loss": 0.9445,
+      "step": 8060
+    },
+    {
+      "epoch": 5.2745098039215685,
+      "grad_norm": 1.329483151435852,
+      "learning_rate": 0.0002,
+      "loss": 0.9724,
+      "step": 8070
+    },
+    {
+      "epoch": 5.281045751633987,
+      "grad_norm": 1.1333061456680298,
+      "learning_rate": 0.0002,
+      "loss": 0.9296,
+      "step": 8080
+    },
+    {
+      "epoch": 5.287581699346405,
+      "grad_norm": 0.8153108358383179,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 8090
+    },
+    {
+      "epoch": 5.294117647058823,
+      "grad_norm": 0.9395004510879517,
+      "learning_rate": 0.0002,
+      "loss": 0.9002,
+      "step": 8100
+    },
+    {
+      "epoch": 5.300653594771242,
+      "grad_norm": 0.8907593488693237,
+      "learning_rate": 0.0002,
+      "loss": 1.0371,
+      "step": 8110
+    },
+    {
+      "epoch": 5.30718954248366,
+      "grad_norm": 0.9808667898178101,
+      "learning_rate": 0.0002,
+      "loss": 0.9301,
+      "step": 8120
+    },
+    {
+      "epoch": 5.313725490196078,
+      "grad_norm": 0.984779417514801,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 8130
+    },
+    {
+      "epoch": 5.3202614379084965,
+      "grad_norm": 0.9787270426750183,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 8140
+    },
+    {
+      "epoch": 5.326797385620915,
+      "grad_norm": 0.9857710599899292,
+      "learning_rate": 0.0002,
+      "loss": 0.9336,
+      "step": 8150
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9774303436279297,
+      "learning_rate": 0.0002,
+      "loss": 0.9884,
+      "step": 8160
+    },
+    {
+      "epoch": 5.339869281045751,
+      "grad_norm": 0.677925169467926,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 8170
+    },
+    {
+      "epoch": 5.34640522875817,
+      "grad_norm": 0.9576456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 8180
+    },
+    {
+      "epoch": 5.352941176470588,
+      "grad_norm": 1.8970937728881836,
+      "learning_rate": 0.0002,
+      "loss": 0.9554,
+      "step": 8190
+    },
+    {
+      "epoch": 5.359477124183006,
+      "grad_norm": 0.9458389282226562,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 8200
+    },
+    {
+      "epoch": 5.366013071895424,
+      "grad_norm": 1.761794924736023,
+      "learning_rate": 0.0002,
+      "loss": 1.0365,
+      "step": 8210
+    },
+    {
+      "epoch": 5.372549019607844,
+      "grad_norm": 1.0693724155426025,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 8220
+    },
+    {
+      "epoch": 5.379084967320262,
+      "grad_norm": 0.9025877714157104,
+      "learning_rate": 0.0002,
+      "loss": 1.0299,
+      "step": 8230
+    },
+    {
+      "epoch": 5.38562091503268,
+      "grad_norm": 1.258857250213623,
+      "learning_rate": 0.0002,
+      "loss": 0.9652,
+      "step": 8240
+    },
+    {
+      "epoch": 5.392156862745098,
+      "grad_norm": 1.084849238395691,
+      "learning_rate": 0.0002,
+      "loss": 0.9735,
+      "step": 8250
+    },
+    {
+      "epoch": 5.398692810457517,
+      "grad_norm": 0.9530340433120728,
+      "learning_rate": 0.0002,
+      "loss": 0.9999,
+      "step": 8260
+    },
+    {
+      "epoch": 5.405228758169935,
+      "grad_norm": 0.830240786075592,
+      "learning_rate": 0.0002,
+      "loss": 1.0268,
+      "step": 8270
+    },
+    {
+      "epoch": 5.411764705882353,
+      "grad_norm": 1.5807015895843506,
+      "learning_rate": 0.0002,
+      "loss": 1.0332,
+      "step": 8280
+    },
+    {
+      "epoch": 5.4183006535947715,
+      "grad_norm": 0.9486905336380005,
+      "learning_rate": 0.0002,
+      "loss": 0.9146,
+      "step": 8290
+    },
+    {
+      "epoch": 5.42483660130719,
+      "grad_norm": 1.0415093898773193,
+      "learning_rate": 0.0002,
+      "loss": 1.0336,
+      "step": 8300
+    },
+    {
+      "epoch": 5.431372549019608,
+      "grad_norm": 1.0501102209091187,
+      "learning_rate": 0.0002,
+      "loss": 0.8933,
+      "step": 8310
+    },
+    {
+      "epoch": 5.437908496732026,
+      "grad_norm": 0.9751836061477661,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 8320
+    },
+    {
+      "epoch": 5.444444444444445,
+      "grad_norm": 1.5529173612594604,
+      "learning_rate": 0.0002,
+      "loss": 1.0755,
+      "step": 8330
+    },
+    {
+      "epoch": 5.450980392156863,
+      "grad_norm": 0.8314350247383118,
+      "learning_rate": 0.0002,
+      "loss": 0.9814,
+      "step": 8340
+    },
+    {
+      "epoch": 5.457516339869281,
+      "grad_norm": 1.2555103302001953,
+      "learning_rate": 0.0002,
+      "loss": 1.0596,
+      "step": 8350
+    },
+    {
+      "epoch": 5.4640522875816995,
+      "grad_norm": 0.9408367872238159,
+      "learning_rate": 0.0002,
+      "loss": 1.0127,
+      "step": 8360
+    },
+    {
+      "epoch": 5.470588235294118,
+      "grad_norm": 0.9483312964439392,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 8370
+    },
+    {
+      "epoch": 5.477124183006536,
+      "grad_norm": 0.957905650138855,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8380
+    },
+    {
+      "epoch": 5.483660130718954,
+      "grad_norm": 1.4000147581100464,
+      "learning_rate": 0.0002,
+      "loss": 1.0985,
+      "step": 8390
+    },
+    {
+      "epoch": 5.490196078431373,
+      "grad_norm": 1.7032461166381836,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 8400
+    },
+    {
+      "epoch": 5.496732026143791,
+      "grad_norm": 0.8978716731071472,
+      "learning_rate": 0.0002,
+      "loss": 0.9539,
+      "step": 8410
+    },
+    {
+      "epoch": 5.503267973856209,
+      "grad_norm": 0.8659300804138184,
+      "learning_rate": 0.0002,
+      "loss": 0.9544,
+      "step": 8420
+    },
+    {
+      "epoch": 5.509803921568627,
+      "grad_norm": 1.3629727363586426,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 8430
+    },
+    {
+      "epoch": 5.516339869281046,
+      "grad_norm": 1.2741984128952026,
+      "learning_rate": 0.0002,
+      "loss": 0.9696,
+      "step": 8440
+    },
+    {
+      "epoch": 5.522875816993464,
+      "grad_norm": 1.3867180347442627,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 8450
+    },
+    {
+      "epoch": 5.529411764705882,
+      "grad_norm": 1.0662001371383667,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 8460
+    },
+    {
+      "epoch": 5.5359477124183005,
+      "grad_norm": 1.7005380392074585,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 8470
+    },
+    {
+      "epoch": 5.542483660130719,
+      "grad_norm": 1.3730385303497314,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 8480
+    },
+    {
+      "epoch": 5.549019607843137,
+      "grad_norm": 1.7737441062927246,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 8490
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.907487690448761,
+      "learning_rate": 0.0002,
+      "loss": 0.9729,
+      "step": 8500
+    },
+    {
+      "epoch": 5.562091503267974,
+      "grad_norm": 0.8882441520690918,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 8510
+    },
+    {
+      "epoch": 5.568627450980392,
+      "grad_norm": 0.8655388951301575,
+      "learning_rate": 0.0002,
+      "loss": 0.973,
+      "step": 8520
+    },
+    {
+      "epoch": 5.57516339869281,
+      "grad_norm": 1.379992961883545,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 8530
+    },
+    {
+      "epoch": 5.5816993464052285,
+      "grad_norm": 1.0021201372146606,
+      "learning_rate": 0.0002,
+      "loss": 1.0174,
+      "step": 8540
+    },
+    {
+      "epoch": 5.588235294117647,
+      "grad_norm": 1.2636926174163818,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 8550
+    },
+    {
+      "epoch": 5.594771241830065,
+      "grad_norm": 1.279025912284851,
+      "learning_rate": 0.0002,
+      "loss": 1.0243,
+      "step": 8560
+    },
+    {
+      "epoch": 5.601307189542483,
+      "grad_norm": 0.8885834217071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9917,
+      "step": 8570
+    },
+    {
+      "epoch": 5.607843137254902,
+      "grad_norm": 1.1975032091140747,
+      "learning_rate": 0.0002,
+      "loss": 0.9849,
+      "step": 8580
+    },
+    {
+      "epoch": 5.61437908496732,
+      "grad_norm": 1.005470871925354,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8590
+    },
+    {
+      "epoch": 5.620915032679738,
+      "grad_norm": 1.104286551475525,
+      "learning_rate": 0.0002,
+      "loss": 0.9947,
+      "step": 8600
+    },
+    {
+      "epoch": 5.627450980392156,
+      "grad_norm": 1.435445785522461,
+      "learning_rate": 0.0002,
+      "loss": 1.0585,
+      "step": 8610
+    },
+    {
+      "epoch": 5.633986928104575,
+      "grad_norm": 1.0270172357559204,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 8620
+    },
+    {
+      "epoch": 5.640522875816993,
+      "grad_norm": 1.0929527282714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 8630
+    },
+    {
+      "epoch": 5.647058823529412,
+      "grad_norm": 1.1061221361160278,
+      "learning_rate": 0.0002,
+      "loss": 0.9694,
+      "step": 8640
+    },
+    {
+      "epoch": 5.65359477124183,
+      "grad_norm": 0.9563149213790894,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 8650
+    },
+    {
+      "epoch": 5.660130718954249,
+      "grad_norm": 1.0434954166412354,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 8660
+    },
+    {
+      "epoch": 5.666666666666667,
+      "grad_norm": 1.3695117235183716,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 8670
+    },
+    {
+      "epoch": 5.673202614379085,
+      "grad_norm": 1.0540564060211182,
+      "learning_rate": 0.0002,
+      "loss": 0.9441,
+      "step": 8680
+    },
+    {
+      "epoch": 5.6797385620915035,
+      "grad_norm": 1.5942492485046387,
+      "learning_rate": 0.0002,
+      "loss": 0.9755,
+      "step": 8690
+    },
+    {
+      "epoch": 5.686274509803922,
+      "grad_norm": 0.9485495090484619,
+      "learning_rate": 0.0002,
+      "loss": 1.0071,
+      "step": 8700
+    },
+    {
+      "epoch": 5.69281045751634,
+      "grad_norm": 1.1483162641525269,
+      "learning_rate": 0.0002,
+      "loss": 0.9998,
+      "step": 8710
+    },
+    {
+      "epoch": 5.699346405228758,
+      "grad_norm": 0.9075471758842468,
+      "learning_rate": 0.0002,
+      "loss": 0.9578,
+      "step": 8720
+    },
+    {
+      "epoch": 5.705882352941177,
+      "grad_norm": 1.7908551692962646,
+      "learning_rate": 0.0002,
+      "loss": 0.9488,
+      "step": 8730
+    },
+    {
+      "epoch": 5.712418300653595,
+      "grad_norm": 0.8867162466049194,
+      "learning_rate": 0.0002,
+      "loss": 1.0163,
+      "step": 8740
+    },
+    {
+      "epoch": 5.718954248366013,
+      "grad_norm": 1.7165148258209229,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 8750
+    },
+    {
+      "epoch": 5.7254901960784315,
+      "grad_norm": 0.9529356956481934,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 8760
+    },
+    {
+      "epoch": 5.73202614379085,
+      "grad_norm": 1.01852548122406,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 8770
+    },
+    {
+      "epoch": 5.738562091503268,
+      "grad_norm": 0.9538423418998718,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 8780
+    },
+    {
+      "epoch": 5.745098039215686,
+      "grad_norm": 0.9007737636566162,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 8790
+    },
+    {
+      "epoch": 5.751633986928105,
+      "grad_norm": 0.9107874035835266,
+      "learning_rate": 0.0002,
+      "loss": 0.9766,
+      "step": 8800
+    },
+    {
+      "epoch": 5.758169934640523,
+      "grad_norm": 0.7379238605499268,
+      "learning_rate": 0.0002,
+      "loss": 0.9212,
+      "step": 8810
+    },
+    {
+      "epoch": 5.764705882352941,
+      "grad_norm": 1.072645902633667,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 8820
+    },
+    {
+      "epoch": 5.771241830065359,
+      "grad_norm": 1.002008080482483,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 8830
+    },
+    {
+      "epoch": 5.777777777777778,
+      "grad_norm": 1.0435924530029297,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 8840
+    },
+    {
+      "epoch": 5.784313725490196,
+      "grad_norm": 0.9874551296234131,
+      "learning_rate": 0.0002,
+      "loss": 0.9458,
+      "step": 8850
+    },
+    {
+      "epoch": 5.790849673202614,
+      "grad_norm": 1.1729662418365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1241,
+      "step": 8860
+    },
+    {
+      "epoch": 5.7973856209150325,
+      "grad_norm": 1.3300775289535522,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 8870
+    },
+    {
+      "epoch": 5.803921568627451,
+      "grad_norm": 1.612707257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.0989,
+      "step": 8880
+    },
+    {
+      "epoch": 5.810457516339869,
+      "grad_norm": 0.9047797322273254,
+      "learning_rate": 0.0002,
+      "loss": 0.9119,
+      "step": 8890
+    },
+    {
+      "epoch": 5.816993464052287,
+      "grad_norm": 1.0958741903305054,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 8900
+    },
+    {
+      "epoch": 5.823529411764706,
+      "grad_norm": 1.0099612474441528,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 8910
+    },
+    {
+      "epoch": 5.830065359477124,
+      "grad_norm": 0.8442328572273254,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 8920
+    },
+    {
+      "epoch": 5.836601307189542,
+      "grad_norm": 1.1388301849365234,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 8930
+    },
+    {
+      "epoch": 5.8431372549019605,
+      "grad_norm": 0.8296026587486267,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 8940
+    },
+    {
+      "epoch": 5.849673202614379,
+      "grad_norm": 1.0843533277511597,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8950
+    },
+    {
+      "epoch": 5.856209150326797,
+      "grad_norm": 0.8496834635734558,
+      "learning_rate": 0.0002,
+      "loss": 1.0009,
+      "step": 8960
+    },
+    {
+      "epoch": 5.862745098039216,
+      "grad_norm": 1.6894690990447998,
+      "learning_rate": 0.0002,
+      "loss": 0.9927,
+      "step": 8970
+    },
+    {
+      "epoch": 5.8692810457516345,
+      "grad_norm": 1.0012282133102417,
+      "learning_rate": 0.0002,
+      "loss": 1.0939,
+      "step": 8980
+    },
+    {
+      "epoch": 5.875816993464053,
+      "grad_norm": 0.8521103262901306,
+      "learning_rate": 0.0002,
+      "loss": 0.9722,
+      "step": 8990
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 1.246841311454773,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 9000
+    },
+    {
+      "epoch": 5.888888888888889,
+      "grad_norm": 0.9941892027854919,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 9010
+    },
+    {
+      "epoch": 5.895424836601308,
+      "grad_norm": 1.067413568496704,
+      "learning_rate": 0.0002,
+      "loss": 0.8754,
+      "step": 9020
+    },
+    {
+      "epoch": 5.901960784313726,
+      "grad_norm": 1.0045088529586792,
+      "learning_rate": 0.0002,
+      "loss": 1.0153,
+      "step": 9030
+    },
+    {
+      "epoch": 5.908496732026144,
+      "grad_norm": 1.383063554763794,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 9040
+    },
+    {
+      "epoch": 5.915032679738562,
+      "grad_norm": 0.8754428625106812,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 9050
+    },
+    {
+      "epoch": 5.921568627450981,
+      "grad_norm": 0.8577388525009155,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 9060
+    },
+    {
+      "epoch": 5.928104575163399,
+      "grad_norm": 0.8718975186347961,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 9070
+    },
+    {
+      "epoch": 5.934640522875817,
+      "grad_norm": 1.1762131452560425,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 9080
+    },
+    {
+      "epoch": 5.9411764705882355,
+      "grad_norm": 1.1025866270065308,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 9090
+    },
+    {
+      "epoch": 5.947712418300654,
+      "grad_norm": 1.0439870357513428,
+      "learning_rate": 0.0002,
+      "loss": 0.9155,
+      "step": 9100
+    },
+    {
+      "epoch": 5.954248366013072,
+      "grad_norm": 1.2411525249481201,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 9110
+    },
+    {
+      "epoch": 5.96078431372549,
+      "grad_norm": 1.0317714214324951,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 9120
+    },
+    {
+      "epoch": 5.967320261437909,
+      "grad_norm": 0.9880492091178894,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 9130
+    },
+    {
+      "epoch": 5.973856209150327,
+      "grad_norm": 0.9039815664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 9140
+    },
+    {
+      "epoch": 5.980392156862745,
+      "grad_norm": 0.9049116373062134,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 9150
+    },
+    {
+      "epoch": 5.9869281045751634,
+      "grad_norm": 0.996749222278595,
+      "learning_rate": 0.0002,
+      "loss": 0.9792,
+      "step": 9160
+    },
+    {
+      "epoch": 5.993464052287582,
+      "grad_norm": 0.8716062307357788,
+      "learning_rate": 0.0002,
+      "loss": 0.8857,
+      "step": 9170
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.3081294298171997,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 9180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.45111083984375,
+      "eval_runtime": 34.7121,
+      "eval_samples_per_second": 12.56,
+      "eval_steps_per_second": 1.584,
+      "step": 9180
+    },
+    {
+      "epoch": 6.006535947712418,
+      "grad_norm": 1.1378029584884644,
+      "learning_rate": 0.0002,
+      "loss": 0.9306,
+      "step": 9190
+    },
+    {
+      "epoch": 6.0130718954248366,
+      "grad_norm": 1.2921233177185059,
+      "learning_rate": 0.0002,
+      "loss": 0.8794,
+      "step": 9200
+    },
+    {
+      "epoch": 6.019607843137255,
+      "grad_norm": 1.039211630821228,
+      "learning_rate": 0.0002,
+      "loss": 0.8145,
+      "step": 9210
+    },
+    {
+      "epoch": 6.026143790849673,
+      "grad_norm": 0.9715196490287781,
+      "learning_rate": 0.0002,
+      "loss": 0.8524,
+      "step": 9220
+    },
+    {
+      "epoch": 6.032679738562091,
+      "grad_norm": 1.220642328262329,
+      "learning_rate": 0.0002,
+      "loss": 1.035,
+      "step": 9230
+    },
+    {
+      "epoch": 6.03921568627451,
+      "grad_norm": 0.854360044002533,
+      "learning_rate": 0.0002,
+      "loss": 0.8468,
+      "step": 9240
+    },
+    {
+      "epoch": 6.045751633986928,
+      "grad_norm": 0.8806933164596558,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 9250
+    },
+    {
+      "epoch": 6.052287581699346,
+      "grad_norm": 1.4315874576568604,
+      "learning_rate": 0.0002,
+      "loss": 0.8305,
+      "step": 9260
+    },
+    {
+      "epoch": 6.0588235294117645,
+      "grad_norm": 0.9382007122039795,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 9270
+    },
+    {
+      "epoch": 6.065359477124183,
+      "grad_norm": 1.2184561491012573,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 9280
+    },
+    {
+      "epoch": 6.071895424836601,
+      "grad_norm": 1.2331548929214478,
+      "learning_rate": 0.0002,
+      "loss": 0.8806,
+      "step": 9290
+    },
+    {
+      "epoch": 6.078431372549019,
+      "grad_norm": 1.1112796068191528,
+      "learning_rate": 0.0002,
+      "loss": 0.8354,
+      "step": 9300
+    },
+    {
+      "epoch": 6.084967320261438,
+      "grad_norm": 1.4753731489181519,
+      "learning_rate": 0.0002,
+      "loss": 0.8008,
+      "step": 9310
+    },
+    {
+      "epoch": 6.091503267973856,
+      "grad_norm": 1.2783401012420654,
+      "learning_rate": 0.0002,
+      "loss": 0.9198,
+      "step": 9320
+    },
+    {
+      "epoch": 6.098039215686274,
+      "grad_norm": 0.9916909337043762,
+      "learning_rate": 0.0002,
+      "loss": 0.8294,
+      "step": 9330
+    },
+    {
+      "epoch": 6.104575163398692,
+      "grad_norm": 0.9300099015235901,
+      "learning_rate": 0.0002,
+      "loss": 0.876,
+      "step": 9340
+    },
+    {
+      "epoch": 6.111111111111111,
+      "grad_norm": 1.4985264539718628,
+      "learning_rate": 0.0002,
+      "loss": 0.9064,
+      "step": 9350
+    },
+    {
+      "epoch": 6.117647058823529,
+      "grad_norm": 1.276380181312561,
+      "learning_rate": 0.0002,
+      "loss": 1.0106,
+      "step": 9360
+    },
+    {
+      "epoch": 6.124183006535947,
+      "grad_norm": 1.181113600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9068,
+      "step": 9370
+    },
+    {
+      "epoch": 6.130718954248366,
+      "grad_norm": 1.698729395866394,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 9380
+    },
+    {
+      "epoch": 6.137254901960785,
+      "grad_norm": 0.9793189764022827,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 9390
+    },
+    {
+      "epoch": 6.143790849673203,
+      "grad_norm": 1.1942132711410522,
+      "learning_rate": 0.0002,
+      "loss": 0.9731,
+      "step": 9400
+    },
+    {
+      "epoch": 6.150326797385621,
+      "grad_norm": 1.2160184383392334,
+      "learning_rate": 0.0002,
+      "loss": 0.8762,
+      "step": 9410
+    },
+    {
+      "epoch": 6.1568627450980395,
+      "grad_norm": 1.0802825689315796,
+      "learning_rate": 0.0002,
+      "loss": 0.801,
+      "step": 9420
+    },
+    {
+      "epoch": 6.163398692810458,
+      "grad_norm": 3.024529218673706,
+      "learning_rate": 0.0002,
+      "loss": 0.9055,
+      "step": 9430
+    },
+    {
+      "epoch": 6.169934640522876,
+      "grad_norm": 0.975062370300293,
+      "learning_rate": 0.0002,
+      "loss": 0.8739,
+      "step": 9440
+    },
+    {
+      "epoch": 6.176470588235294,
+      "grad_norm": 0.9243306517601013,
+      "learning_rate": 0.0002,
+      "loss": 0.8485,
+      "step": 9450
+    },
+    {
+      "epoch": 6.183006535947713,
+      "grad_norm": 0.8892099857330322,
+      "learning_rate": 0.0002,
+      "loss": 0.947,
+      "step": 9460
+    },
+    {
+      "epoch": 6.189542483660131,
+      "grad_norm": 1.4151731729507446,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 9470
+    },
+    {
+      "epoch": 6.196078431372549,
+      "grad_norm": 1.064701795578003,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 9480
+    },
+    {
+      "epoch": 6.2026143790849675,
+      "grad_norm": 1.1104519367218018,
+      "learning_rate": 0.0002,
+      "loss": 0.906,
+      "step": 9490
+    },
+    {
+      "epoch": 6.209150326797386,
+      "grad_norm": 1.4788947105407715,
+      "learning_rate": 0.0002,
+      "loss": 0.9572,
+      "step": 9500
+    },
+    {
+      "epoch": 6.215686274509804,
+      "grad_norm": 0.7976077795028687,
+      "learning_rate": 0.0002,
+      "loss": 0.8014,
+      "step": 9510
+    },
+    {
+      "epoch": 6.222222222222222,
+      "grad_norm": 1.256864070892334,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 9520
+    },
+    {
+      "epoch": 6.228758169934641,
+      "grad_norm": 1.3874554634094238,
+      "learning_rate": 0.0002,
+      "loss": 0.9104,
+      "step": 9530
+    },
+    {
+      "epoch": 6.235294117647059,
+      "grad_norm": 1.9012963771820068,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 9540
+    },
+    {
+      "epoch": 6.241830065359477,
+      "grad_norm": 1.275212287902832,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 9550
+    },
+    {
+      "epoch": 6.248366013071895,
+      "grad_norm": 1.1007417440414429,
+      "learning_rate": 0.0002,
+      "loss": 0.8416,
+      "step": 9560
+    },
+    {
+      "epoch": 6.254901960784314,
+      "grad_norm": 1.0602147579193115,
+      "learning_rate": 0.0002,
+      "loss": 0.9191,
+      "step": 9570
+    },
+    {
+      "epoch": 6.261437908496732,
+      "grad_norm": 1.2276418209075928,
+      "learning_rate": 0.0002,
+      "loss": 0.909,
+      "step": 9580
+    },
+    {
+      "epoch": 6.26797385620915,
+      "grad_norm": 1.0111924409866333,
+      "learning_rate": 0.0002,
+      "loss": 0.9363,
+      "step": 9590
+    },
+    {
+      "epoch": 6.2745098039215685,
+      "grad_norm": 0.9031485915184021,
+      "learning_rate": 0.0002,
+      "loss": 0.9941,
+      "step": 9600
+    },
+    {
+      "epoch": 6.281045751633987,
+      "grad_norm": 0.9893783926963806,
+      "learning_rate": 0.0002,
+      "loss": 0.9138,
+      "step": 9610
+    },
+    {
+      "epoch": 6.287581699346405,
+      "grad_norm": 1.1979725360870361,
+      "learning_rate": 0.0002,
+      "loss": 0.9114,
+      "step": 9620
+    },
+    {
+      "epoch": 6.294117647058823,
+      "grad_norm": 1.380516767501831,
+      "learning_rate": 0.0002,
+      "loss": 0.8858,
+      "step": 9630
+    },
+    {
+      "epoch": 6.300653594771242,
+      "grad_norm": 1.1370083093643188,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 9640
+    },
+    {
+      "epoch": 6.30718954248366,
+      "grad_norm": 1.4091558456420898,
+      "learning_rate": 0.0002,
+      "loss": 0.9073,
+      "step": 9650
+    },
+    {
+      "epoch": 6.313725490196078,
+      "grad_norm": 1.0670944452285767,
+      "learning_rate": 0.0002,
+      "loss": 0.9096,
+      "step": 9660
+    },
+    {
+      "epoch": 6.3202614379084965,
+      "grad_norm": 0.9150263667106628,
+      "learning_rate": 0.0002,
+      "loss": 0.9376,
+      "step": 9670
+    },
+    {
+      "epoch": 6.326797385620915,
+      "grad_norm": 1.1342853307724,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 9680
+    },
+    {
+      "epoch": 6.333333333333333,
+      "grad_norm": 1.2733415365219116,
+      "learning_rate": 0.0002,
+      "loss": 1.002,
+      "step": 9690
+    },
+    {
+      "epoch": 6.339869281045751,
+      "grad_norm": 1.3647292852401733,
+      "learning_rate": 0.0002,
+      "loss": 0.9579,
+      "step": 9700
+    },
+    {
+      "epoch": 6.34640522875817,
+      "grad_norm": 1.0435094833374023,
+      "learning_rate": 0.0002,
+      "loss": 0.87,
+      "step": 9710
+    },
+    {
+      "epoch": 6.352941176470588,
+      "grad_norm": 1.3641071319580078,
+      "learning_rate": 0.0002,
+      "loss": 0.8812,
+      "step": 9720
+    },
+    {
+      "epoch": 6.359477124183006,
+      "grad_norm": 1.2806159257888794,
+      "learning_rate": 0.0002,
+      "loss": 0.8888,
+      "step": 9730
+    },
+    {
+      "epoch": 6.366013071895424,
+      "grad_norm": 1.0193076133728027,
+      "learning_rate": 0.0002,
+      "loss": 0.9481,
+      "step": 9740
+    },
+    {
+      "epoch": 6.372549019607844,
+      "grad_norm": 1.2349408864974976,
+      "learning_rate": 0.0002,
+      "loss": 0.931,
+      "step": 9750
+    },
+    {
+      "epoch": 6.379084967320262,
+      "grad_norm": 1.2062549591064453,
+      "learning_rate": 0.0002,
+      "loss": 0.8837,
+      "step": 9760
+    },
+    {
+      "epoch": 6.38562091503268,
+      "grad_norm": 1.4402194023132324,
+      "learning_rate": 0.0002,
+      "loss": 0.8947,
+      "step": 9770
+    },
+    {
+      "epoch": 6.392156862745098,
+      "grad_norm": 1.1730891466140747,
+      "learning_rate": 0.0002,
+      "loss": 0.8724,
+      "step": 9780
+    },
+    {
+      "epoch": 6.398692810457517,
+      "grad_norm": 1.1481093168258667,
+      "learning_rate": 0.0002,
+      "loss": 0.9005,
+      "step": 9790
+    },
+    {
+      "epoch": 6.405228758169935,
+      "grad_norm": 1.0012723207473755,
+      "learning_rate": 0.0002,
+      "loss": 0.9431,
+      "step": 9800
+    },
+    {
+      "epoch": 6.411764705882353,
+      "grad_norm": 0.8839848041534424,
+      "learning_rate": 0.0002,
+      "loss": 0.8856,
+      "step": 9810
+    },
+    {
+      "epoch": 6.4183006535947715,
+      "grad_norm": 1.096693992614746,
+      "learning_rate": 0.0002,
+      "loss": 0.8147,
+      "step": 9820
+    },
+    {
+      "epoch": 6.42483660130719,
+      "grad_norm": 1.4713369607925415,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 9830
+    },
+    {
+      "epoch": 6.431372549019608,
+      "grad_norm": 1.2529761791229248,
+      "learning_rate": 0.0002,
+      "loss": 0.9563,
+      "step": 9840
+    },
+    {
+      "epoch": 6.437908496732026,
+      "grad_norm": 1.5575600862503052,
+      "learning_rate": 0.0002,
+      "loss": 0.8551,
+      "step": 9850
+    },
+    {
+      "epoch": 6.444444444444445,
+      "grad_norm": 1.2188916206359863,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 9860
+    },
+    {
+      "epoch": 6.450980392156863,
+      "grad_norm": 1.1558794975280762,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 9870
+    },
+    {
+      "epoch": 6.457516339869281,
+      "grad_norm": 1.1506937742233276,
+      "learning_rate": 0.0002,
+      "loss": 0.8632,
+      "step": 9880
+    },
+    {
+      "epoch": 6.4640522875816995,
+      "grad_norm": 1.1168335676193237,
+      "learning_rate": 0.0002,
+      "loss": 1.0575,
+      "step": 9890
+    },
+    {
+      "epoch": 6.470588235294118,
+      "grad_norm": 1.192449688911438,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 9900
+    },
+    {
+      "epoch": 6.477124183006536,
+      "grad_norm": 1.0451104640960693,
+      "learning_rate": 0.0002,
+      "loss": 0.9478,
+      "step": 9910
+    },
+    {
+      "epoch": 6.483660130718954,
+      "grad_norm": 1.1111775636672974,
+      "learning_rate": 0.0002,
+      "loss": 0.9034,
+      "step": 9920
+    },
+    {
+      "epoch": 6.490196078431373,
+      "grad_norm": 1.2094531059265137,
+      "learning_rate": 0.0002,
+      "loss": 0.8971,
+      "step": 9930
+    },
+    {
+      "epoch": 6.496732026143791,
+      "grad_norm": 1.0547380447387695,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 9940
+    },
+    {
+      "epoch": 6.503267973856209,
+      "grad_norm": 1.5547202825546265,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 9950
+    },
+    {
+      "epoch": 6.509803921568627,
+      "grad_norm": 1.1917903423309326,
+      "learning_rate": 0.0002,
+      "loss": 0.9109,
+      "step": 9960
+    },
+    {
+      "epoch": 6.516339869281046,
+      "grad_norm": 1.0918153524398804,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 9970
+    },
+    {
+      "epoch": 6.522875816993464,
+      "grad_norm": 1.146968960762024,
+      "learning_rate": 0.0002,
+      "loss": 0.8752,
+      "step": 9980
+    },
+    {
+      "epoch": 6.529411764705882,
+      "grad_norm": 0.9899234771728516,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 9990
+    },
+    {
+      "epoch": 6.5359477124183005,
+      "grad_norm": 2.160924196243286,
+      "learning_rate": 0.0002,
+      "loss": 0.91,
+      "step": 10000
+    },
+    {
+      "epoch": 6.542483660130719,
+      "grad_norm": 1.6366891860961914,
+      "learning_rate": 0.0002,
+      "loss": 0.9683,
+      "step": 10010
+    },
+    {
+      "epoch": 6.549019607843137,
+      "grad_norm": 0.9876762628555298,
+      "learning_rate": 0.0002,
+      "loss": 0.8582,
+      "step": 10020
+    },
+    {
+      "epoch": 6.555555555555555,
+      "grad_norm": 1.5622549057006836,
+      "learning_rate": 0.0002,
+      "loss": 0.8385,
+      "step": 10030
+    },
+    {
+      "epoch": 6.562091503267974,
+      "grad_norm": 1.0108020305633545,
+      "learning_rate": 0.0002,
+      "loss": 0.8791,
+      "step": 10040
+    },
+    {
+      "epoch": 6.568627450980392,
+      "grad_norm": 1.0725725889205933,
+      "learning_rate": 0.0002,
+      "loss": 0.9574,
+      "step": 10050
+    },
+    {
+      "epoch": 6.57516339869281,
+      "grad_norm": 1.1551216840744019,
+      "learning_rate": 0.0002,
+      "loss": 0.8297,
+      "step": 10060
+    },
+    {
+      "epoch": 6.5816993464052285,
+      "grad_norm": 1.5174646377563477,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 10070
+    },
+    {
+      "epoch": 6.588235294117647,
+      "grad_norm": 1.041877031326294,
+      "learning_rate": 0.0002,
+      "loss": 0.8203,
+      "step": 10080
+    },
+    {
+      "epoch": 6.594771241830065,
+      "grad_norm": 0.9939621686935425,
+      "learning_rate": 0.0002,
+      "loss": 0.9684,
+      "step": 10090
+    },
+    {
+      "epoch": 6.601307189542483,
+      "grad_norm": 1.2706589698791504,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 10100
+    },
+    {
+      "epoch": 6.607843137254902,
+      "grad_norm": 1.1071467399597168,
+      "learning_rate": 0.0002,
+      "loss": 0.9614,
+      "step": 10110
+    },
+    {
+      "epoch": 6.61437908496732,
+      "grad_norm": 0.9449541568756104,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 10120
+    },
+    {
+      "epoch": 6.620915032679738,
+      "grad_norm": 1.0961830615997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9557,
+      "step": 10130
+    },
+    {
+      "epoch": 6.627450980392156,
+      "grad_norm": 1.7726300954818726,
+      "learning_rate": 0.0002,
+      "loss": 0.9865,
+      "step": 10140
+    },
+    {
+      "epoch": 6.633986928104575,
+      "grad_norm": 1.2345516681671143,
+      "learning_rate": 0.0002,
+      "loss": 0.9657,
+      "step": 10150
+    },
+    {
+      "epoch": 6.640522875816993,
+      "grad_norm": 1.2062907218933105,
+      "learning_rate": 0.0002,
+      "loss": 0.9573,
+      "step": 10160
+    },
+    {
+      "epoch": 6.647058823529412,
+      "grad_norm": 1.029327154159546,
+      "learning_rate": 0.0002,
+      "loss": 0.918,
+      "step": 10170
+    },
+    {
+      "epoch": 6.65359477124183,
+      "grad_norm": 1.442307710647583,
+      "learning_rate": 0.0002,
+      "loss": 0.9211,
+      "step": 10180
+    },
+    {
+      "epoch": 6.660130718954249,
+      "grad_norm": 1.2579066753387451,
+      "learning_rate": 0.0002,
+      "loss": 0.8924,
+      "step": 10190
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.4563188552856445,
+      "learning_rate": 0.0002,
+      "loss": 0.9836,
+      "step": 10200
+    },
+    {
+      "epoch": 6.673202614379085,
+      "grad_norm": 0.9699450135231018,
+      "learning_rate": 0.0002,
+      "loss": 0.8876,
+      "step": 10210
+    },
+    {
+      "epoch": 6.6797385620915035,
+      "grad_norm": 1.812523603439331,
+      "learning_rate": 0.0002,
+      "loss": 0.9589,
+      "step": 10220
+    },
+    {
+      "epoch": 6.686274509803922,
+      "grad_norm": 1.124000906944275,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 10230
+    },
+    {
+      "epoch": 6.69281045751634,
+      "grad_norm": 1.0957475900650024,
+      "learning_rate": 0.0002,
+      "loss": 0.8924,
+      "step": 10240
+    },
+    {
+      "epoch": 6.699346405228758,
+      "grad_norm": 0.989689826965332,
+      "learning_rate": 0.0002,
+      "loss": 0.8891,
+      "step": 10250
+    },
+    {
+      "epoch": 6.705882352941177,
+      "grad_norm": 1.4353317022323608,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 10260
+    },
+    {
+      "epoch": 6.712418300653595,
+      "grad_norm": 1.0245451927185059,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 10270
+    },
+    {
+      "epoch": 6.718954248366013,
+      "grad_norm": 1.097334861755371,
+      "learning_rate": 0.0002,
+      "loss": 0.8814,
+      "step": 10280
+    },
+    {
+      "epoch": 6.7254901960784315,
+      "grad_norm": 0.982356071472168,
+      "learning_rate": 0.0002,
+      "loss": 0.9927,
+      "step": 10290
+    },
+    {
+      "epoch": 6.73202614379085,
+      "grad_norm": 1.8842819929122925,
+      "learning_rate": 0.0002,
+      "loss": 0.9909,
+      "step": 10300
+    },
+    {
+      "epoch": 6.738562091503268,
+      "grad_norm": 0.8648947477340698,
+      "learning_rate": 0.0002,
+      "loss": 0.9286,
+      "step": 10310
+    },
+    {
+      "epoch": 6.745098039215686,
+      "grad_norm": 1.1510577201843262,
+      "learning_rate": 0.0002,
+      "loss": 0.987,
+      "step": 10320
+    },
+    {
+      "epoch": 6.751633986928105,
+      "grad_norm": 1.874495506286621,
+      "learning_rate": 0.0002,
+      "loss": 0.9217,
+      "step": 10330
+    },
+    {
+      "epoch": 6.758169934640523,
+      "grad_norm": 1.1126408576965332,
+      "learning_rate": 0.0002,
+      "loss": 0.8914,
+      "step": 10340
+    },
+    {
+      "epoch": 6.764705882352941,
+      "grad_norm": 1.6654644012451172,
+      "learning_rate": 0.0002,
+      "loss": 0.8508,
+      "step": 10350
+    },
+    {
+      "epoch": 6.771241830065359,
+      "grad_norm": 1.0699580907821655,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 10360
+    },
+    {
+      "epoch": 6.777777777777778,
+      "grad_norm": 0.9460757374763489,
+      "learning_rate": 0.0002,
+      "loss": 0.882,
+      "step": 10370
+    },
+    {
+      "epoch": 6.784313725490196,
+      "grad_norm": 1.2553058862686157,
+      "learning_rate": 0.0002,
+      "loss": 0.9589,
+      "step": 10380
+    },
+    {
+      "epoch": 6.790849673202614,
+      "grad_norm": 1.0939891338348389,
+      "learning_rate": 0.0002,
+      "loss": 0.8782,
+      "step": 10390
+    },
+    {
+      "epoch": 6.7973856209150325,
+      "grad_norm": 1.0647451877593994,
+      "learning_rate": 0.0002,
+      "loss": 0.9189,
+      "step": 10400
+    },
+    {
+      "epoch": 6.803921568627451,
+      "grad_norm": 1.0954521894454956,
+      "learning_rate": 0.0002,
+      "loss": 0.9478,
+      "step": 10410
+    },
+    {
+      "epoch": 6.810457516339869,
+      "grad_norm": 1.4371392726898193,
+      "learning_rate": 0.0002,
+      "loss": 1.0385,
+      "step": 10420
+    },
+    {
+      "epoch": 6.816993464052287,
+      "grad_norm": 1.0063464641571045,
+      "learning_rate": 0.0002,
+      "loss": 1.0024,
+      "step": 10430
+    },
+    {
+      "epoch": 6.823529411764706,
+      "grad_norm": 1.5189263820648193,
+      "learning_rate": 0.0002,
+      "loss": 0.8737,
+      "step": 10440
+    },
+    {
+      "epoch": 6.830065359477124,
+      "grad_norm": 0.9715501070022583,
+      "learning_rate": 0.0002,
+      "loss": 0.9246,
+      "step": 10450
+    },
+    {
+      "epoch": 6.836601307189542,
+      "grad_norm": 1.114586353302002,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 10460
+    },
+    {
+      "epoch": 6.8431372549019605,
+      "grad_norm": 1.2991431951522827,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 10470
+    },
+    {
+      "epoch": 6.849673202614379,
+      "grad_norm": 1.203114628791809,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 10480
+    },
+    {
+      "epoch": 6.856209150326797,
+      "grad_norm": 1.476167917251587,
+      "learning_rate": 0.0002,
+      "loss": 1.0032,
+      "step": 10490
+    },
+    {
+      "epoch": 6.862745098039216,
+      "grad_norm": 1.0933326482772827,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 10500
+    },
+    {
+      "epoch": 6.8692810457516345,
+      "grad_norm": 1.2831504344940186,
+      "learning_rate": 0.0002,
+      "loss": 1.0068,
+      "step": 10510
+    },
+    {
+      "epoch": 6.875816993464053,
+      "grad_norm": 1.1967637538909912,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 10520
+    },
+    {
+      "epoch": 6.882352941176471,
+      "grad_norm": 1.1276888847351074,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 10530
+    },
+    {
+      "epoch": 6.888888888888889,
+      "grad_norm": 1.2680490016937256,
+      "learning_rate": 0.0002,
+      "loss": 0.9568,
+      "step": 10540
+    },
+    {
+      "epoch": 6.895424836601308,
+      "grad_norm": 1.5469038486480713,
+      "learning_rate": 0.0002,
+      "loss": 0.9177,
+      "step": 10550
+    },
+    {
+      "epoch": 6.901960784313726,
+      "grad_norm": 1.1731038093566895,
+      "learning_rate": 0.0002,
+      "loss": 0.8545,
+      "step": 10560
+    },
+    {
+      "epoch": 6.908496732026144,
+      "grad_norm": 0.968008816242218,
+      "learning_rate": 0.0002,
+      "loss": 0.9795,
+      "step": 10570
+    },
+    {
+      "epoch": 6.915032679738562,
+      "grad_norm": 0.9082416892051697,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 10580
+    },
+    {
+      "epoch": 6.921568627450981,
+      "grad_norm": 1.5816899538040161,
+      "learning_rate": 0.0002,
+      "loss": 0.9898,
+      "step": 10590
+    },
+    {
+      "epoch": 6.928104575163399,
+      "grad_norm": 0.9462234377861023,
+      "learning_rate": 0.0002,
+      "loss": 0.9692,
+      "step": 10600
+    },
+    {
+      "epoch": 6.934640522875817,
+      "grad_norm": 1.4950200319290161,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 10610
+    },
+    {
+      "epoch": 6.9411764705882355,
+      "grad_norm": 1.2929182052612305,
+      "learning_rate": 0.0002,
+      "loss": 0.8888,
+      "step": 10620
+    },
+    {
+      "epoch": 6.947712418300654,
+      "grad_norm": 1.2995754480361938,
+      "learning_rate": 0.0002,
+      "loss": 1.0141,
+      "step": 10630
+    },
+    {
+      "epoch": 6.954248366013072,
+      "grad_norm": 0.9407122135162354,
+      "learning_rate": 0.0002,
+      "loss": 0.9863,
+      "step": 10640
+    },
+    {
+      "epoch": 6.96078431372549,
+      "grad_norm": 1.1735378503799438,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 10650
+    },
+    {
+      "epoch": 6.967320261437909,
+      "grad_norm": 0.9937344193458557,
+      "learning_rate": 0.0002,
+      "loss": 0.936,
+      "step": 10660
+    },
+    {
+      "epoch": 6.973856209150327,
+      "grad_norm": 1.2498728036880493,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 10670
+    },
+    {
+      "epoch": 6.980392156862745,
+      "grad_norm": 1.0513341426849365,
+      "learning_rate": 0.0002,
+      "loss": 1.0504,
+      "step": 10680
+    },
+    {
+      "epoch": 6.9869281045751634,
+      "grad_norm": 1.4611467123031616,
+      "learning_rate": 0.0002,
+      "loss": 0.9259,
+      "step": 10690
+    },
+    {
+      "epoch": 6.993464052287582,
+      "grad_norm": 1.2924799919128418,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 10700
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.2024929523468018,
+      "learning_rate": 0.0002,
+      "loss": 0.8953,
+      "step": 10710
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.4972445964813232,
+      "eval_runtime": 33.6225,
+      "eval_samples_per_second": 12.967,
+      "eval_steps_per_second": 1.636,
+      "step": 10710
+    },
+    {
+      "epoch": 7.006535947712418,
+      "grad_norm": 1.1302162408828735,
+      "learning_rate": 0.0002,
+      "loss": 0.8111,
+      "step": 10720
+    },
+    {
+      "epoch": 7.0130718954248366,
+      "grad_norm": 1.2731552124023438,
+      "learning_rate": 0.0002,
+      "loss": 0.8584,
+      "step": 10730
+    },
+    {
+      "epoch": 7.019607843137255,
+      "grad_norm": 1.2694480419158936,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 10740
+    },
+    {
+      "epoch": 7.026143790849673,
+      "grad_norm": 1.1517360210418701,
+      "learning_rate": 0.0002,
+      "loss": 0.7621,
+      "step": 10750
+    },
+    {
+      "epoch": 7.032679738562091,
+      "grad_norm": 1.3649171590805054,
+      "learning_rate": 0.0002,
+      "loss": 0.9149,
+      "step": 10760
+    },
+    {
+      "epoch": 7.03921568627451,
+      "grad_norm": 1.1630656719207764,
+      "learning_rate": 0.0002,
+      "loss": 0.7411,
+      "step": 10770
+    },
+    {
+      "epoch": 7.045751633986928,
+      "grad_norm": 1.2658313512802124,
+      "learning_rate": 0.0002,
+      "loss": 0.8514,
+      "step": 10780
+    },
+    {
+      "epoch": 7.052287581699346,
+      "grad_norm": 1.5004769563674927,
+      "learning_rate": 0.0002,
+      "loss": 0.8125,
+      "step": 10790
+    },
+    {
+      "epoch": 7.0588235294117645,
+      "grad_norm": 1.052678108215332,
+      "learning_rate": 0.0002,
+      "loss": 0.754,
+      "step": 10800
+    },
+    {
+      "epoch": 7.065359477124183,
+      "grad_norm": 1.3461277484893799,
+      "learning_rate": 0.0002,
+      "loss": 0.8613,
+      "step": 10810
+    },
+    {
+      "epoch": 7.071895424836601,
+      "grad_norm": 1.3074769973754883,
+      "learning_rate": 0.0002,
+      "loss": 0.843,
+      "step": 10820
+    },
+    {
+      "epoch": 7.078431372549019,
+      "grad_norm": 1.8454785346984863,
+      "learning_rate": 0.0002,
+      "loss": 0.7433,
+      "step": 10830
+    },
+    {
+      "epoch": 7.084967320261438,
+      "grad_norm": 0.9786653518676758,
+      "learning_rate": 0.0002,
+      "loss": 0.7899,
+      "step": 10840
+    },
+    {
+      "epoch": 7.091503267973856,
+      "grad_norm": 1.2760838270187378,
+      "learning_rate": 0.0002,
+      "loss": 0.7689,
+      "step": 10850
+    },
+    {
+      "epoch": 7.098039215686274,
+      "grad_norm": 1.1340841054916382,
+      "learning_rate": 0.0002,
+      "loss": 0.7715,
+      "step": 10860
+    },
+    {
+      "epoch": 7.104575163398692,
+      "grad_norm": 1.3808159828186035,
+      "learning_rate": 0.0002,
+      "loss": 0.8111,
+      "step": 10870
+    },
+    {
+      "epoch": 7.111111111111111,
+      "grad_norm": 1.147668719291687,
+      "learning_rate": 0.0002,
+      "loss": 0.8311,
+      "step": 10880
+    },
+    {
+      "epoch": 7.117647058823529,
+      "grad_norm": 1.3183035850524902,
+      "learning_rate": 0.0002,
+      "loss": 0.8408,
+      "step": 10890
+    },
+    {
+      "epoch": 7.124183006535947,
+      "grad_norm": 1.3882936239242554,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 10900
+    },
+    {
+      "epoch": 7.130718954248366,
+      "grad_norm": 0.9495398998260498,
+      "learning_rate": 0.0002,
+      "loss": 0.7745,
+      "step": 10910
+    },
+    {
+      "epoch": 7.137254901960785,
+      "grad_norm": 1.3810124397277832,
+      "learning_rate": 0.0002,
+      "loss": 0.8438,
+      "step": 10920
+    },
+    {
+      "epoch": 7.143790849673203,
+      "grad_norm": 1.563207745552063,
+      "learning_rate": 0.0002,
+      "loss": 0.8028,
+      "step": 10930
+    },
+    {
+      "epoch": 7.150326797385621,
+      "grad_norm": 1.2633056640625,
+      "learning_rate": 0.0002,
+      "loss": 0.7562,
+      "step": 10940
+    },
+    {
+      "epoch": 7.1568627450980395,
+      "grad_norm": 1.2398860454559326,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 10950
+    },
+    {
+      "epoch": 7.163398692810458,
+      "grad_norm": 1.166763186454773,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 10960
+    },
+    {
+      "epoch": 7.169934640522876,
+      "grad_norm": 1.5083234310150146,
+      "learning_rate": 0.0002,
+      "loss": 0.8346,
+      "step": 10970
+    },
+    {
+      "epoch": 7.176470588235294,
+      "grad_norm": 1.6927601099014282,
+      "learning_rate": 0.0002,
+      "loss": 0.8323,
+      "step": 10980
+    },
+    {
+      "epoch": 7.183006535947713,
+      "grad_norm": 1.090780258178711,
+      "learning_rate": 0.0002,
+      "loss": 0.7434,
+      "step": 10990
+    },
+    {
+      "epoch": 7.189542483660131,
+      "grad_norm": 1.0077793598175049,
+      "learning_rate": 0.0002,
+      "loss": 0.784,
+      "step": 11000
+    },
+    {
+      "epoch": 7.196078431372549,
+      "grad_norm": 1.8293051719665527,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 11010
+    },
+    {
+      "epoch": 7.2026143790849675,
+      "grad_norm": 1.0761457681655884,
+      "learning_rate": 0.0002,
+      "loss": 0.7654,
+      "step": 11020
+    },
+    {
+      "epoch": 7.209150326797386,
+      "grad_norm": 1.0681469440460205,
+      "learning_rate": 0.0002,
+      "loss": 0.8395,
+      "step": 11030
+    },
+    {
+      "epoch": 7.215686274509804,
+      "grad_norm": 1.961199402809143,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 11040
+    },
+    {
+      "epoch": 7.222222222222222,
+      "grad_norm": 1.3750165700912476,
+      "learning_rate": 0.0002,
+      "loss": 0.8631,
+      "step": 11050
+    },
+    {
+      "epoch": 7.228758169934641,
+      "grad_norm": 1.647005319595337,
+      "learning_rate": 0.0002,
+      "loss": 0.7425,
+      "step": 11060
+    },
+    {
+      "epoch": 7.235294117647059,
+      "grad_norm": 1.1073668003082275,
+      "learning_rate": 0.0002,
+      "loss": 0.844,
+      "step": 11070
+    },
+    {
+      "epoch": 7.241830065359477,
+      "grad_norm": 1.450289011001587,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 11080
+    },
+    {
+      "epoch": 7.248366013071895,
+      "grad_norm": 1.191163420677185,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 11090
+    },
+    {
+      "epoch": 7.254901960784314,
+      "grad_norm": 1.6975404024124146,
+      "learning_rate": 0.0002,
+      "loss": 0.8331,
+      "step": 11100
+    },
+    {
+      "epoch": 7.261437908496732,
+      "grad_norm": 1.159091830253601,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 11110
+    },
+    {
+      "epoch": 7.26797385620915,
+      "grad_norm": 0.9952927827835083,
+      "learning_rate": 0.0002,
+      "loss": 0.8612,
+      "step": 11120
+    },
+    {
+      "epoch": 7.2745098039215685,
+      "grad_norm": 1.4122034311294556,
+      "learning_rate": 0.0002,
+      "loss": 0.8299,
+      "step": 11130
+    },
+    {
+      "epoch": 7.281045751633987,
+      "grad_norm": 1.4299325942993164,
+      "learning_rate": 0.0002,
+      "loss": 0.8274,
+      "step": 11140
+    },
+    {
+      "epoch": 7.287581699346405,
+      "grad_norm": 1.26812744140625,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 11150
+    },
+    {
+      "epoch": 7.294117647058823,
+      "grad_norm": 1.0740736722946167,
+      "learning_rate": 0.0002,
+      "loss": 0.8036,
+      "step": 11160
+    },
+    {
+      "epoch": 7.300653594771242,
+      "grad_norm": 1.1293542385101318,
+      "learning_rate": 0.0002,
+      "loss": 0.9005,
+      "step": 11170
+    },
+    {
+      "epoch": 7.30718954248366,
+      "grad_norm": 1.3161042928695679,
+      "learning_rate": 0.0002,
+      "loss": 0.8775,
+      "step": 11180
+    },
+    {
+      "epoch": 7.313725490196078,
+      "grad_norm": 1.5637391805648804,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 11190
+    },
+    {
+      "epoch": 7.3202614379084965,
+      "grad_norm": 1.3164077997207642,
+      "learning_rate": 0.0002,
+      "loss": 0.8704,
+      "step": 11200
+    },
+    {
+      "epoch": 7.326797385620915,
+      "grad_norm": 0.9268870949745178,
+      "learning_rate": 0.0002,
+      "loss": 0.8552,
+      "step": 11210
+    },
+    {
+      "epoch": 7.333333333333333,
+      "grad_norm": 1.164515733718872,
+      "learning_rate": 0.0002,
+      "loss": 0.8719,
+      "step": 11220
+    },
+    {
+      "epoch": 7.339869281045751,
+      "grad_norm": 1.6878753900527954,
+      "learning_rate": 0.0002,
+      "loss": 0.9011,
+      "step": 11230
+    },
+    {
+      "epoch": 7.34640522875817,
+      "grad_norm": 1.1870672702789307,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 11240
+    },
+    {
+      "epoch": 7.352941176470588,
+      "grad_norm": 1.2923716306686401,
+      "learning_rate": 0.0002,
+      "loss": 0.886,
+      "step": 11250
+    },
+    {
+      "epoch": 7.359477124183006,
+      "grad_norm": 1.2006791830062866,
+      "learning_rate": 0.0002,
+      "loss": 0.7763,
+      "step": 11260
+    },
+    {
+      "epoch": 7.366013071895424,
+      "grad_norm": 1.0424097776412964,
+      "learning_rate": 0.0002,
+      "loss": 0.7859,
+      "step": 11270
+    },
+    {
+      "epoch": 7.372549019607844,
+      "grad_norm": 1.349094033241272,
+      "learning_rate": 0.0002,
+      "loss": 0.9427,
+      "step": 11280
+    },
+    {
+      "epoch": 7.379084967320262,
+      "grad_norm": 1.4128005504608154,
+      "learning_rate": 0.0002,
+      "loss": 0.8384,
+      "step": 11290
+    },
+    {
+      "epoch": 7.38562091503268,
+      "grad_norm": 1.3647041320800781,
+      "learning_rate": 0.0002,
+      "loss": 0.9219,
+      "step": 11300
+    },
+    {
+      "epoch": 7.392156862745098,
+      "grad_norm": 1.0561704635620117,
+      "learning_rate": 0.0002,
+      "loss": 0.7865,
+      "step": 11310
+    },
+    {
+      "epoch": 7.398692810457517,
+      "grad_norm": 1.2405760288238525,
+      "learning_rate": 0.0002,
+      "loss": 0.9151,
+      "step": 11320
+    },
+    {
+      "epoch": 7.405228758169935,
+      "grad_norm": 1.0932328701019287,
+      "learning_rate": 0.0002,
+      "loss": 0.9108,
+      "step": 11330
+    },
+    {
+      "epoch": 7.411764705882353,
+      "grad_norm": 1.206778883934021,
+      "learning_rate": 0.0002,
+      "loss": 0.873,
+      "step": 11340
+    },
+    {
+      "epoch": 7.4183006535947715,
+      "grad_norm": 1.5261255502700806,
+      "learning_rate": 0.0002,
+      "loss": 0.9062,
+      "step": 11350
+    },
+    {
+      "epoch": 7.42483660130719,
+      "grad_norm": 1.4928070306777954,
+      "learning_rate": 0.0002,
+      "loss": 0.7795,
+      "step": 11360
+    },
+    {
+      "epoch": 7.431372549019608,
+      "grad_norm": 1.331190586090088,
+      "learning_rate": 0.0002,
+      "loss": 0.9027,
+      "step": 11370
+    },
+    {
+      "epoch": 7.437908496732026,
+      "grad_norm": 1.0745981931686401,
+      "learning_rate": 0.0002,
+      "loss": 0.8547,
+      "step": 11380
+    },
+    {
+      "epoch": 7.444444444444445,
+      "grad_norm": 1.3070036172866821,
+      "learning_rate": 0.0002,
+      "loss": 0.7755,
+      "step": 11390
+    },
+    {
+      "epoch": 7.450980392156863,
+      "grad_norm": 0.8743805885314941,
+      "learning_rate": 0.0002,
+      "loss": 0.7703,
+      "step": 11400
+    },
+    {
+      "epoch": 7.457516339869281,
+      "grad_norm": 1.2747994661331177,
+      "learning_rate": 0.0002,
+      "loss": 0.9331,
+      "step": 11410
+    },
+    {
+      "epoch": 7.4640522875816995,
+      "grad_norm": 1.3688995838165283,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 11420
+    },
+    {
+      "epoch": 7.470588235294118,
+      "grad_norm": 1.1788195371627808,
+      "learning_rate": 0.0002,
+      "loss": 0.783,
+      "step": 11430
+    },
+    {
+      "epoch": 7.477124183006536,
+      "grad_norm": 2.0186705589294434,
+      "learning_rate": 0.0002,
+      "loss": 0.8841,
+      "step": 11440
+    },
+    {
+      "epoch": 7.483660130718954,
+      "grad_norm": 1.1707696914672852,
+      "learning_rate": 0.0002,
+      "loss": 0.9182,
+      "step": 11450
+    },
+    {
+      "epoch": 7.490196078431373,
+      "grad_norm": 1.26426100730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9019,
+      "step": 11460
+    },
+    {
+      "epoch": 7.496732026143791,
+      "grad_norm": 1.2673691511154175,
+      "learning_rate": 0.0002,
+      "loss": 0.8114,
+      "step": 11470
+    },
+    {
+      "epoch": 7.503267973856209,
+      "grad_norm": 1.038956642150879,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 11480
+    },
+    {
+      "epoch": 7.509803921568627,
+      "grad_norm": 1.216252326965332,
+      "learning_rate": 0.0002,
+      "loss": 0.8066,
+      "step": 11490
+    },
+    {
+      "epoch": 7.516339869281046,
+      "grad_norm": 1.1520167589187622,
+      "learning_rate": 0.0002,
+      "loss": 0.8546,
+      "step": 11500
+    },
+    {
+      "epoch": 7.522875816993464,
+      "grad_norm": 1.3962451219558716,
+      "learning_rate": 0.0002,
+      "loss": 0.8117,
+      "step": 11510
+    },
+    {
+      "epoch": 7.529411764705882,
+      "grad_norm": 1.2226953506469727,
+      "learning_rate": 0.0002,
+      "loss": 0.8636,
+      "step": 11520
+    },
+    {
+      "epoch": 7.5359477124183005,
+      "grad_norm": 1.2891474962234497,
+      "learning_rate": 0.0002,
+      "loss": 0.8943,
+      "step": 11530
+    },
+    {
+      "epoch": 7.542483660130719,
+      "grad_norm": 1.3372766971588135,
+      "learning_rate": 0.0002,
+      "loss": 0.8787,
+      "step": 11540
+    },
+    {
+      "epoch": 7.549019607843137,
+      "grad_norm": 1.4196370840072632,
+      "learning_rate": 0.0002,
+      "loss": 0.8103,
+      "step": 11550
+    },
+    {
+      "epoch": 7.555555555555555,
+      "grad_norm": 1.0041396617889404,
+      "learning_rate": 0.0002,
+      "loss": 0.8722,
+      "step": 11560
+    },
+    {
+      "epoch": 7.562091503267974,
+      "grad_norm": 1.3470606803894043,
+      "learning_rate": 0.0002,
+      "loss": 0.8512,
+      "step": 11570
+    },
+    {
+      "epoch": 7.568627450980392,
+      "grad_norm": 1.1738601922988892,
+      "learning_rate": 0.0002,
+      "loss": 0.8472,
+      "step": 11580
+    },
+    {
+      "epoch": 7.57516339869281,
+      "grad_norm": 1.1629133224487305,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 11590
+    },
+    {
+      "epoch": 7.5816993464052285,
+      "grad_norm": 1.2859786748886108,
+      "learning_rate": 0.0002,
+      "loss": 0.8502,
+      "step": 11600
+    },
+    {
+      "epoch": 7.588235294117647,
+      "grad_norm": 1.429398775100708,
+      "learning_rate": 0.0002,
+      "loss": 0.8356,
+      "step": 11610
+    },
+    {
+      "epoch": 7.594771241830065,
+      "grad_norm": 1.3300801515579224,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 11620
+    },
+    {
+      "epoch": 7.601307189542483,
+      "grad_norm": 1.3261592388153076,
+      "learning_rate": 0.0002,
+      "loss": 0.8671,
+      "step": 11630
+    },
+    {
+      "epoch": 7.607843137254902,
+      "grad_norm": 1.8779925107955933,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 11640
+    },
+    {
+      "epoch": 7.61437908496732,
+      "grad_norm": 1.7839158773422241,
+      "learning_rate": 0.0002,
+      "loss": 0.7666,
+      "step": 11650
+    },
+    {
+      "epoch": 7.620915032679738,
+      "grad_norm": 1.6469435691833496,
+      "learning_rate": 0.0002,
+      "loss": 0.9163,
+      "step": 11660
+    },
+    {
+      "epoch": 7.627450980392156,
+      "grad_norm": 1.5416018962860107,
+      "learning_rate": 0.0002,
+      "loss": 0.8589,
+      "step": 11670
+    },
+    {
+      "epoch": 7.633986928104575,
+      "grad_norm": 1.5173335075378418,
+      "learning_rate": 0.0002,
+      "loss": 0.7112,
+      "step": 11680
+    },
+    {
+      "epoch": 7.640522875816993,
+      "grad_norm": 1.1372658014297485,
+      "learning_rate": 0.0002,
+      "loss": 0.8958,
+      "step": 11690
+    },
+    {
+      "epoch": 7.647058823529412,
+      "grad_norm": 1.233030915260315,
+      "learning_rate": 0.0002,
+      "loss": 0.8852,
+      "step": 11700
+    },
+    {
+      "epoch": 7.65359477124183,
+      "grad_norm": 1.3100069761276245,
+      "learning_rate": 0.0002,
+      "loss": 0.9205,
+      "step": 11710
+    },
+    {
+      "epoch": 7.660130718954249,
+      "grad_norm": 0.9770023226737976,
+      "learning_rate": 0.0002,
+      "loss": 0.8895,
+      "step": 11720
+    },
+    {
+      "epoch": 7.666666666666667,
+      "grad_norm": 1.240946650505066,
+      "learning_rate": 0.0002,
+      "loss": 0.8658,
+      "step": 11730
+    },
+    {
+      "epoch": 7.673202614379085,
+      "grad_norm": 1.444226861000061,
+      "learning_rate": 0.0002,
+      "loss": 0.9527,
+      "step": 11740
+    },
+    {
+      "epoch": 7.6797385620915035,
+      "grad_norm": 1.2667231559753418,
+      "learning_rate": 0.0002,
+      "loss": 0.8448,
+      "step": 11750
+    },
+    {
+      "epoch": 7.686274509803922,
+      "grad_norm": 1.340754747390747,
+      "learning_rate": 0.0002,
+      "loss": 0.9195,
+      "step": 11760
+    },
+    {
+      "epoch": 7.69281045751634,
+      "grad_norm": 1.181988000869751,
+      "learning_rate": 0.0002,
+      "loss": 0.8829,
+      "step": 11770
+    },
+    {
+      "epoch": 7.699346405228758,
+      "grad_norm": 1.0623301267623901,
+      "learning_rate": 0.0002,
+      "loss": 0.8609,
+      "step": 11780
+    },
+    {
+      "epoch": 7.705882352941177,
+      "grad_norm": 1.1917353868484497,
+      "learning_rate": 0.0002,
+      "loss": 0.7726,
+      "step": 11790
+    },
+    {
+      "epoch": 7.712418300653595,
+      "grad_norm": 1.7202110290527344,
+      "learning_rate": 0.0002,
+      "loss": 0.8075,
+      "step": 11800
+    },
+    {
+      "epoch": 7.718954248366013,
+      "grad_norm": 1.1121439933776855,
+      "learning_rate": 0.0002,
+      "loss": 0.8112,
+      "step": 11810
+    },
+    {
+      "epoch": 7.7254901960784315,
+      "grad_norm": 0.956794261932373,
+      "learning_rate": 0.0002,
+      "loss": 0.7831,
+      "step": 11820
+    },
+    {
+      "epoch": 7.73202614379085,
+      "grad_norm": 1.2524380683898926,
+      "learning_rate": 0.0002,
+      "loss": 0.9056,
+      "step": 11830
+    },
+    {
+      "epoch": 7.738562091503268,
+      "grad_norm": 1.1095308065414429,
+      "learning_rate": 0.0002,
+      "loss": 0.8337,
+      "step": 11840
+    },
+    {
+      "epoch": 7.745098039215686,
+      "grad_norm": 1.631195068359375,
+      "learning_rate": 0.0002,
+      "loss": 0.8921,
+      "step": 11850
+    },
+    {
+      "epoch": 7.751633986928105,
+      "grad_norm": 1.2265965938568115,
+      "learning_rate": 0.0002,
+      "loss": 0.8493,
+      "step": 11860
+    },
+    {
+      "epoch": 7.758169934640523,
+      "grad_norm": 1.080328106880188,
+      "learning_rate": 0.0002,
+      "loss": 0.8875,
+      "step": 11870
+    },
+    {
+      "epoch": 7.764705882352941,
+      "grad_norm": 1.5570356845855713,
+      "learning_rate": 0.0002,
+      "loss": 0.8732,
+      "step": 11880
+    },
+    {
+      "epoch": 7.771241830065359,
+      "grad_norm": 1.3791661262512207,
+      "learning_rate": 0.0002,
+      "loss": 0.8293,
+      "step": 11890
+    },
+    {
+      "epoch": 7.777777777777778,
+      "grad_norm": 1.1457891464233398,
+      "learning_rate": 0.0002,
+      "loss": 0.8333,
+      "step": 11900
+    },
+    {
+      "epoch": 7.784313725490196,
+      "grad_norm": 1.6357585191726685,
+      "learning_rate": 0.0002,
+      "loss": 0.8546,
+      "step": 11910
+    },
+    {
+      "epoch": 7.790849673202614,
+      "grad_norm": 1.1845953464508057,
+      "learning_rate": 0.0002,
+      "loss": 0.9041,
+      "step": 11920
+    },
+    {
+      "epoch": 7.7973856209150325,
+      "grad_norm": 1.2255016565322876,
+      "learning_rate": 0.0002,
+      "loss": 0.8364,
+      "step": 11930
+    },
+    {
+      "epoch": 7.803921568627451,
+      "grad_norm": 1.2113513946533203,
+      "learning_rate": 0.0002,
+      "loss": 0.913,
+      "step": 11940
+    },
+    {
+      "epoch": 7.810457516339869,
+      "grad_norm": 1.0834609270095825,
+      "learning_rate": 0.0002,
+      "loss": 0.8553,
+      "step": 11950
+    },
+    {
+      "epoch": 7.816993464052287,
+      "grad_norm": 1.0127689838409424,
+      "learning_rate": 0.0002,
+      "loss": 0.8125,
+      "step": 11960
+    },
+    {
+      "epoch": 7.823529411764706,
+      "grad_norm": 1.1124000549316406,
+      "learning_rate": 0.0002,
+      "loss": 0.9468,
+      "step": 11970
+    },
+    {
+      "epoch": 7.830065359477124,
+      "grad_norm": 1.3440804481506348,
+      "learning_rate": 0.0002,
+      "loss": 0.8345,
+      "step": 11980
+    },
+    {
+      "epoch": 7.836601307189542,
+      "grad_norm": 1.8478741645812988,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 11990
+    },
+    {
+      "epoch": 7.8431372549019605,
+      "grad_norm": 1.1202499866485596,
+      "learning_rate": 0.0002,
+      "loss": 0.9708,
+      "step": 12000
+    },
+    {
+      "epoch": 7.849673202614379,
+      "grad_norm": 1.735700249671936,
+      "learning_rate": 0.0002,
+      "loss": 0.8437,
+      "step": 12010
+    },
+    {
+      "epoch": 7.856209150326797,
+      "grad_norm": 1.2994014024734497,
+      "learning_rate": 0.0002,
+      "loss": 0.8333,
+      "step": 12020
+    },
+    {
+      "epoch": 7.862745098039216,
+      "grad_norm": 1.8655444383621216,
+      "learning_rate": 0.0002,
+      "loss": 0.8656,
+      "step": 12030
+    },
+    {
+      "epoch": 7.8692810457516345,
+      "grad_norm": 1.0460877418518066,
+      "learning_rate": 0.0002,
+      "loss": 0.8919,
+      "step": 12040
+    },
+    {
+      "epoch": 7.875816993464053,
+      "grad_norm": 1.5241339206695557,
+      "learning_rate": 0.0002,
+      "loss": 0.8603,
+      "step": 12050
+    },
+    {
+      "epoch": 7.882352941176471,
+      "grad_norm": 1.171849250793457,
+      "learning_rate": 0.0002,
+      "loss": 0.8179,
+      "step": 12060
+    },
+    {
+      "epoch": 7.888888888888889,
+      "grad_norm": 1.0957022905349731,
+      "learning_rate": 0.0002,
+      "loss": 0.8577,
+      "step": 12070
+    },
+    {
+      "epoch": 7.895424836601308,
+      "grad_norm": 1.4121248722076416,
+      "learning_rate": 0.0002,
+      "loss": 0.9212,
+      "step": 12080
+    },
+    {
+      "epoch": 7.901960784313726,
+      "grad_norm": 1.3393208980560303,
+      "learning_rate": 0.0002,
+      "loss": 1.0002,
+      "step": 12090
+    },
+    {
+      "epoch": 7.908496732026144,
+      "grad_norm": 1.1252245903015137,
+      "learning_rate": 0.0002,
+      "loss": 0.8959,
+      "step": 12100
+    },
+    {
+      "epoch": 7.915032679738562,
+      "grad_norm": 1.4131813049316406,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 12110
+    },
+    {
+      "epoch": 7.921568627450981,
+      "grad_norm": 1.2392992973327637,
+      "learning_rate": 0.0002,
+      "loss": 0.9106,
+      "step": 12120
+    },
+    {
+      "epoch": 7.928104575163399,
+      "grad_norm": 1.3233672380447388,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 12130
+    },
+    {
+      "epoch": 7.934640522875817,
+      "grad_norm": 1.2547026872634888,
+      "learning_rate": 0.0002,
+      "loss": 0.7853,
+      "step": 12140
+    },
+    {
+      "epoch": 7.9411764705882355,
+      "grad_norm": 1.1143239736557007,
+      "learning_rate": 0.0002,
+      "loss": 0.9203,
+      "step": 12150
+    },
+    {
+      "epoch": 7.947712418300654,
+      "grad_norm": 1.030006766319275,
+      "learning_rate": 0.0002,
+      "loss": 0.8059,
+      "step": 12160
+    },
+    {
+      "epoch": 7.954248366013072,
+      "grad_norm": 1.1070104837417603,
+      "learning_rate": 0.0002,
+      "loss": 0.8076,
+      "step": 12170
+    },
+    {
+      "epoch": 7.96078431372549,
+      "grad_norm": 1.3011643886566162,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 12180
+    },
+    {
+      "epoch": 7.967320261437909,
+      "grad_norm": 1.134848713874817,
+      "learning_rate": 0.0002,
+      "loss": 0.7951,
+      "step": 12190
+    },
+    {
+      "epoch": 7.973856209150327,
+      "grad_norm": 1.7021794319152832,
+      "learning_rate": 0.0002,
+      "loss": 0.9318,
+      "step": 12200
+    },
+    {
+      "epoch": 7.980392156862745,
+      "grad_norm": 1.0190330743789673,
+      "learning_rate": 0.0002,
+      "loss": 0.9159,
+      "step": 12210
+    },
+    {
+      "epoch": 7.9869281045751634,
+      "grad_norm": 1.6083006858825684,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 12220
+    },
+    {
+      "epoch": 7.993464052287582,
+      "grad_norm": 0.8929536938667297,
+      "learning_rate": 0.0002,
+      "loss": 0.915,
+      "step": 12230
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.9928004145622253,
+      "learning_rate": 0.0002,
+      "loss": 0.8706,
+      "step": 12240
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.5263545513153076,
+      "eval_runtime": 33.6089,
+      "eval_samples_per_second": 12.973,
+      "eval_steps_per_second": 1.636,
+      "step": 12240
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4959723514167296e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-12240/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4306725fab3f86257413add00072b452f5b8e372
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f778dcff85f98db7eeda468dace1f52104faf8349ee5c62b0084a871aeab7db5
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d4a22974d92a2f259ec999275da99537188677ed
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:055b61904899d8c245e186619c3ee84593b4e93b7250d5443736b1ea7ae816c8
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fbc4dbfb6f1d07cc9580cb72fab1022550017edf
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f430b73c5984667b50d1408aabb054172543e00726abc905e82bd691bb3ed14
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..12cf840f119983494dc57dcded908b0a8a439bd4
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70fd0e651657c68255bfbddaea06e66fc5308abd761799251251b4b129a8c90d
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b0e8f96a945315844fd7ee0f6d67a270e120d4c2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/trainer_state.json
@@ -0,0 +1,1112 @@
+{
+  "best_metric": 1.4715123176574707,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530",
+  "epoch": 1.0,
+  "eval_steps": 10,
+  "global_step": 1530,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 1.869965439270912e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..542e836189e0394a10bf371313fbebcaf90101d9
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da6c2b53ce5155b34304e447a4ffda48feee6f707478ea05cbbc61bf87bee38d
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..303805891fb1481cdbbf73f8b79bbfe4b19ba6a3
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a545166f9c84bea18e36eff1c076f0eb9f0cdecace8ff6bfde27df9751de2a
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d7876af9ab6f2d52f56d1b3311ccf18f023466fa
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ee770ffc0dc82e97c84094c424d42999df56b9c94b5bb2694bb76f9c86374b5
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb86c1901ca58fffe630b9be02cf95d9158d1e63
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c601eee32e7508f647875ff3375d30f0d39da2c21bb1297b36d7da7ee8d423c0
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8bf67207ff92467519d066c4cea922cc8ce70012
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/trainer_state.json
@@ -0,0 +1,2191 @@
+{
+  "best_metric": 1.4276371002197266,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 3060,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 3.739930878541824e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4ac84c222b4db7e08b3a542e902c817c21e78452
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35003eda35920d58097bfa5b4a6bb46aece58f29cfbfc46af1f150b8e4bb942b
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..393ef27287048b1a7957d989bd487390fade0a27
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f4b257e5f9941c554b614aacbe478f3e231b8140ed9e4e435b36137f35e0d13
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3ba248fd3fb977b0cf0facc03a2cb1f3487feea3
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cf91e502b853d3b5a87bb99bdf09477e284e18af1eb51f31bb51f5b038e5c17
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..062f5f1ebaf4d2445357abb82b2270ca561fa77d
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:390683d95e3073976e31c984bd808b3e2c9a840f1f23bfdd2e7ab3e19fa5c49f
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8e56e4dad19cb386005d50380f15e2106717beef
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/trainer_state.json
@@ -0,0 +1,3270 @@
+{
+  "best_metric": 1.4131312370300293,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590",
+  "epoch": 3.0,
+  "eval_steps": 10,
+  "global_step": 4590,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 5.609896317812736e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9a25de9298ae67ee0c6a132a503d40bed8ca8c54
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0e46139fbdbe7c9dba8da0dc7642049d7fa284650682279a059b4fd5f0e13ba
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d6d918a481b717f232ee93f55effbcd545f64e2b
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6245e8d34d6bddc9e342779575e283d303ea9f42c30bb851955babfbdc58fbe
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4dd5144f46607e90c6f38c8be1df28fdc445669e
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0014af00868b92f66ddbd9e54f88c848b226c17b47294e56eadd924245e83166
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c98e036e7354dc6e66ea46ccbcc385851151283f
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1ba7db4a140f522050ecce2a409dc22f2e6b188c5b8eac264d1c306ae2a65f4
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..f56d642e7625a6eda9f34f9583ec3e2acf0919d3
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/trainer_state.json
@@ -0,0 +1,4349 @@
+{
+  "best_metric": 1.4113320112228394,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 6120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0065359477124183,
+      "grad_norm": 0.6926420331001282,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0130718954248366,
+      "grad_norm": 0.6656355857849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 4610
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 0.9901936650276184,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 4620
+    },
+    {
+      "epoch": 3.026143790849673,
+      "grad_norm": 0.6713474988937378,
+      "learning_rate": 0.0002,
+      "loss": 1.22,
+      "step": 4630
+    },
+    {
+      "epoch": 3.0326797385620914,
+      "grad_norm": 0.6199324131011963,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 4640
+    },
+    {
+      "epoch": 3.0392156862745097,
+      "grad_norm": 0.7180785536766052,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 4650
+    },
+    {
+      "epoch": 3.045751633986928,
+      "grad_norm": 0.8256588578224182,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 4660
+    },
+    {
+      "epoch": 3.052287581699346,
+      "grad_norm": 0.6637389063835144,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 4670
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 0.6980698108673096,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 4680
+    },
+    {
+      "epoch": 3.065359477124183,
+      "grad_norm": 0.8091534972190857,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 4690
+    },
+    {
+      "epoch": 3.0718954248366015,
+      "grad_norm": 0.5715174078941345,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 4700
+    },
+    {
+      "epoch": 3.0784313725490198,
+      "grad_norm": 0.735639750957489,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 4710
+    },
+    {
+      "epoch": 3.084967320261438,
+      "grad_norm": 0.7619708180427551,
+      "learning_rate": 0.0002,
+      "loss": 1.1522,
+      "step": 4720
+    },
+    {
+      "epoch": 3.0915032679738563,
+      "grad_norm": 1.263566017150879,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 4730
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 0.6600871682167053,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4740
+    },
+    {
+      "epoch": 3.104575163398693,
+      "grad_norm": 0.717792809009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 4750
+    },
+    {
+      "epoch": 3.111111111111111,
+      "grad_norm": 0.853714644908905,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1176470588235294,
+      "grad_norm": 1.1004153490066528,
+      "learning_rate": 0.0002,
+      "loss": 1.2031,
+      "step": 4770
+    },
+    {
+      "epoch": 3.1241830065359477,
+      "grad_norm": 0.8566235899925232,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 4780
+    },
+    {
+      "epoch": 3.130718954248366,
+      "grad_norm": 0.8315296173095703,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 4790
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 0.8020524978637695,
+      "learning_rate": 0.0002,
+      "loss": 1.32,
+      "step": 4800
+    },
+    {
+      "epoch": 3.1437908496732025,
+      "grad_norm": 0.7564275860786438,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 4810
+    },
+    {
+      "epoch": 3.150326797385621,
+      "grad_norm": 0.9077776670455933,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 4820
+    },
+    {
+      "epoch": 3.156862745098039,
+      "grad_norm": 0.6323099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 4830
+    },
+    {
+      "epoch": 3.1633986928104574,
+      "grad_norm": 0.6625368595123291,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 4840
+    },
+    {
+      "epoch": 3.1699346405228757,
+      "grad_norm": 0.8119261860847473,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 4850
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 0.6399450898170471,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 4860
+    },
+    {
+      "epoch": 3.183006535947712,
+      "grad_norm": 1.0659016370773315,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 4870
+    },
+    {
+      "epoch": 3.189542483660131,
+      "grad_norm": 0.8040369749069214,
+      "learning_rate": 0.0002,
+      "loss": 1.2914,
+      "step": 4880
+    },
+    {
+      "epoch": 3.196078431372549,
+      "grad_norm": 0.7784733176231384,
+      "learning_rate": 0.0002,
+      "loss": 1.1996,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2026143790849675,
+      "grad_norm": 0.9660294651985168,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 4900
+    },
+    {
+      "epoch": 3.2091503267973858,
+      "grad_norm": 1.0676977634429932,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 4910
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 0.5877565741539001,
+      "learning_rate": 0.0002,
+      "loss": 1.0083,
+      "step": 4920
+    },
+    {
+      "epoch": 3.2222222222222223,
+      "grad_norm": 0.6164032816886902,
+      "learning_rate": 0.0002,
+      "loss": 1.1046,
+      "step": 4930
+    },
+    {
+      "epoch": 3.2287581699346406,
+      "grad_norm": 0.7627606987953186,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 4940
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.7442803978919983,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 4950
+    },
+    {
+      "epoch": 3.241830065359477,
+      "grad_norm": 0.7277812361717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 4960
+    },
+    {
+      "epoch": 3.2483660130718954,
+      "grad_norm": 1.0301902294158936,
+      "learning_rate": 0.0002,
+      "loss": 1.2237,
+      "step": 4970
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 0.7798232436180115,
+      "learning_rate": 0.0002,
+      "loss": 1.1466,
+      "step": 4980
+    },
+    {
+      "epoch": 3.261437908496732,
+      "grad_norm": 1.210265874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.2142,
+      "step": 4990
+    },
+    {
+      "epoch": 3.2679738562091503,
+      "grad_norm": 0.6677713990211487,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2745098039215685,
+      "grad_norm": 1.0524500608444214,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 5010
+    },
+    {
+      "epoch": 3.281045751633987,
+      "grad_norm": 0.7091745734214783,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5020
+    },
+    {
+      "epoch": 3.287581699346405,
+      "grad_norm": 0.8523224592208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1891,
+      "step": 5030
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 0.6120608448982239,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3006535947712417,
+      "grad_norm": 0.7437472939491272,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 5050
+    },
+    {
+      "epoch": 3.30718954248366,
+      "grad_norm": 0.7611715197563171,
+      "learning_rate": 0.0002,
+      "loss": 1.1295,
+      "step": 5060
+    },
+    {
+      "epoch": 3.313725490196078,
+      "grad_norm": 0.7249704003334045,
+      "learning_rate": 0.0002,
+      "loss": 1.0531,
+      "step": 5070
+    },
+    {
+      "epoch": 3.3202614379084965,
+      "grad_norm": 0.7316247820854187,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5080
+    },
+    {
+      "epoch": 3.326797385620915,
+      "grad_norm": 0.562412440776825,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5090
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.7052176594734192,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5100
+    },
+    {
+      "epoch": 3.3398692810457518,
+      "grad_norm": 0.7714211344718933,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 5110
+    },
+    {
+      "epoch": 3.34640522875817,
+      "grad_norm": 1.0436055660247803,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5120
+    },
+    {
+      "epoch": 3.3529411764705883,
+      "grad_norm": 0.8867271542549133,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 5130
+    },
+    {
+      "epoch": 3.3594771241830066,
+      "grad_norm": 0.8371267914772034,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 5140
+    },
+    {
+      "epoch": 3.366013071895425,
+      "grad_norm": 0.7257837057113647,
+      "learning_rate": 0.0002,
+      "loss": 1.1073,
+      "step": 5150
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 0.7102002501487732,
+      "learning_rate": 0.0002,
+      "loss": 1.1162,
+      "step": 5160
+    },
+    {
+      "epoch": 3.3790849673202614,
+      "grad_norm": 0.7636350393295288,
+      "learning_rate": 0.0002,
+      "loss": 1.2056,
+      "step": 5170
+    },
+    {
+      "epoch": 3.3856209150326797,
+      "grad_norm": 0.6887359619140625,
+      "learning_rate": 0.0002,
+      "loss": 1.0708,
+      "step": 5180
+    },
+    {
+      "epoch": 3.392156862745098,
+      "grad_norm": 0.8141424655914307,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 5190
+    },
+    {
+      "epoch": 3.3986928104575163,
+      "grad_norm": 0.694423496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5200
+    },
+    {
+      "epoch": 3.4052287581699345,
+      "grad_norm": 0.914013683795929,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5210
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 0.8503239750862122,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 5220
+    },
+    {
+      "epoch": 3.418300653594771,
+      "grad_norm": 0.6196836233139038,
+      "learning_rate": 0.0002,
+      "loss": 1.2696,
+      "step": 5230
+    },
+    {
+      "epoch": 3.4248366013071894,
+      "grad_norm": 1.0760811567306519,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 5240
+    },
+    {
+      "epoch": 3.431372549019608,
+      "grad_norm": 0.6524698138237,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 5250
+    },
+    {
+      "epoch": 3.4379084967320264,
+      "grad_norm": 0.674467921257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5260
+    },
+    {
+      "epoch": 3.4444444444444446,
+      "grad_norm": 0.7690372467041016,
+      "learning_rate": 0.0002,
+      "loss": 1.1015,
+      "step": 5270
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 0.8751813769340515,
+      "learning_rate": 0.0002,
+      "loss": 1.2511,
+      "step": 5280
+    },
+    {
+      "epoch": 3.457516339869281,
+      "grad_norm": 0.750407874584198,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 5290
+    },
+    {
+      "epoch": 3.4640522875816995,
+      "grad_norm": 0.5991823077201843,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 5300
+    },
+    {
+      "epoch": 3.4705882352941178,
+      "grad_norm": 1.0164772272109985,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 5310
+    },
+    {
+      "epoch": 3.477124183006536,
+      "grad_norm": 0.8704105019569397,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 5320
+    },
+    {
+      "epoch": 3.4836601307189543,
+      "grad_norm": 0.709102213382721,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 5330
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 0.6273632049560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 5340
+    },
+    {
+      "epoch": 3.496732026143791,
+      "grad_norm": 0.6807359457015991,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 5350
+    },
+    {
+      "epoch": 3.503267973856209,
+      "grad_norm": 0.7085188627243042,
+      "learning_rate": 0.0002,
+      "loss": 1.131,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5098039215686274,
+      "grad_norm": 0.6938307881355286,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 5370
+    },
+    {
+      "epoch": 3.5163398692810457,
+      "grad_norm": 0.8544146418571472,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 5380
+    },
+    {
+      "epoch": 3.522875816993464,
+      "grad_norm": 0.7889642119407654,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 5390
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.7858421206474304,
+      "learning_rate": 0.0002,
+      "loss": 1.1691,
+      "step": 5400
+    },
+    {
+      "epoch": 3.5359477124183005,
+      "grad_norm": 0.8547123074531555,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5410
+    },
+    {
+      "epoch": 3.542483660130719,
+      "grad_norm": 0.8218181133270264,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 5420
+    },
+    {
+      "epoch": 3.549019607843137,
+      "grad_norm": 1.153623342514038,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 5430
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.1321099996566772,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 5440
+    },
+    {
+      "epoch": 3.5620915032679736,
+      "grad_norm": 0.9495334029197693,
+      "learning_rate": 0.0002,
+      "loss": 1.2224,
+      "step": 5450
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 0.8743821978569031,
+      "learning_rate": 0.0002,
+      "loss": 1.2869,
+      "step": 5460
+    },
+    {
+      "epoch": 3.57516339869281,
+      "grad_norm": 0.7513086795806885,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 5470
+    },
+    {
+      "epoch": 3.581699346405229,
+      "grad_norm": 1.0139480829238892,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 5480
+    },
+    {
+      "epoch": 3.588235294117647,
+      "grad_norm": 0.6615135073661804,
+      "learning_rate": 0.0002,
+      "loss": 1.1706,
+      "step": 5490
+    },
+    {
+      "epoch": 3.5947712418300655,
+      "grad_norm": 1.180798888206482,
+      "learning_rate": 0.0002,
+      "loss": 1.3906,
+      "step": 5500
+    },
+    {
+      "epoch": 3.6013071895424837,
+      "grad_norm": 0.7085279226303101,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 5510
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 0.540268063545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1623,
+      "step": 5520
+    },
+    {
+      "epoch": 3.6143790849673203,
+      "grad_norm": 0.7905671000480652,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 5530
+    },
+    {
+      "epoch": 3.6209150326797386,
+      "grad_norm": 0.8457717299461365,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 5540
+    },
+    {
+      "epoch": 3.627450980392157,
+      "grad_norm": 0.7102677822113037,
+      "learning_rate": 0.0002,
+      "loss": 1.1799,
+      "step": 5550
+    },
+    {
+      "epoch": 3.633986928104575,
+      "grad_norm": 0.7179514765739441,
+      "learning_rate": 0.0002,
+      "loss": 1.2394,
+      "step": 5560
+    },
+    {
+      "epoch": 3.6405228758169934,
+      "grad_norm": 1.0854148864746094,
+      "learning_rate": 0.0002,
+      "loss": 1.2019,
+      "step": 5570
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 0.8209951519966125,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5580
+    },
+    {
+      "epoch": 3.65359477124183,
+      "grad_norm": 0.6944138407707214,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 5590
+    },
+    {
+      "epoch": 3.6601307189542482,
+      "grad_norm": 0.7675473093986511,
+      "learning_rate": 0.0002,
+      "loss": 1.3226,
+      "step": 5600
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6683364510536194,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 5610
+    },
+    {
+      "epoch": 3.6732026143790852,
+      "grad_norm": 0.7920727133750916,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 5620
+    },
+    {
+      "epoch": 3.6797385620915035,
+      "grad_norm": 0.9440218806266785,
+      "learning_rate": 0.0002,
+      "loss": 1.2287,
+      "step": 5630
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 0.6600824594497681,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 5640
+    },
+    {
+      "epoch": 3.69281045751634,
+      "grad_norm": 0.6860619187355042,
+      "learning_rate": 0.0002,
+      "loss": 1.191,
+      "step": 5650
+    },
+    {
+      "epoch": 3.6993464052287583,
+      "grad_norm": 0.6579713225364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 5660
+    },
+    {
+      "epoch": 3.7058823529411766,
+      "grad_norm": 0.661081075668335,
+      "learning_rate": 0.0002,
+      "loss": 1.1464,
+      "step": 5670
+    },
+    {
+      "epoch": 3.712418300653595,
+      "grad_norm": 1.0968825817108154,
+      "learning_rate": 0.0002,
+      "loss": 1.289,
+      "step": 5680
+    },
+    {
+      "epoch": 3.718954248366013,
+      "grad_norm": 0.8066844940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.192,
+      "step": 5690
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 0.8341682553291321,
+      "learning_rate": 0.0002,
+      "loss": 1.2322,
+      "step": 5700
+    },
+    {
+      "epoch": 3.7320261437908497,
+      "grad_norm": 0.6682852506637573,
+      "learning_rate": 0.0002,
+      "loss": 1.1473,
+      "step": 5710
+    },
+    {
+      "epoch": 3.738562091503268,
+      "grad_norm": 0.898595929145813,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 5720
+    },
+    {
+      "epoch": 3.7450980392156863,
+      "grad_norm": 0.6876054406166077,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 5730
+    },
+    {
+      "epoch": 3.7516339869281046,
+      "grad_norm": 0.7817103266716003,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 5740
+    },
+    {
+      "epoch": 3.758169934640523,
+      "grad_norm": 0.5840168595314026,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 5750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 0.6263918876647949,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5760
+    },
+    {
+      "epoch": 3.7712418300653594,
+      "grad_norm": 0.7948952317237854,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 5770
+    },
+    {
+      "epoch": 3.7777777777777777,
+      "grad_norm": 0.6700998544692993,
+      "learning_rate": 0.0002,
+      "loss": 1.149,
+      "step": 5780
+    },
+    {
+      "epoch": 3.784313725490196,
+      "grad_norm": 1.1169519424438477,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 5790
+    },
+    {
+      "epoch": 3.7908496732026142,
+      "grad_norm": 0.8354471325874329,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 5800
+    },
+    {
+      "epoch": 3.7973856209150325,
+      "grad_norm": 0.6304181814193726,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 5810
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 0.6919655799865723,
+      "learning_rate": 0.0002,
+      "loss": 1.2059,
+      "step": 5820
+    },
+    {
+      "epoch": 3.810457516339869,
+      "grad_norm": 0.600385844707489,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5830
+    },
+    {
+      "epoch": 3.8169934640522873,
+      "grad_norm": 0.8406319618225098,
+      "learning_rate": 0.0002,
+      "loss": 1.2324,
+      "step": 5840
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.7594282031059265,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5850
+    },
+    {
+      "epoch": 3.8300653594771243,
+      "grad_norm": 0.8179879784584045,
+      "learning_rate": 0.0002,
+      "loss": 1.1903,
+      "step": 5860
+    },
+    {
+      "epoch": 3.8366013071895426,
+      "grad_norm": 1.141430377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 5870
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 0.6595550775527954,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 5880
+    },
+    {
+      "epoch": 3.849673202614379,
+      "grad_norm": 0.7499435544013977,
+      "learning_rate": 0.0002,
+      "loss": 1.2378,
+      "step": 5890
+    },
+    {
+      "epoch": 3.8562091503267975,
+      "grad_norm": 0.7851517200469971,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5900
+    },
+    {
+      "epoch": 3.8627450980392157,
+      "grad_norm": 1.0533545017242432,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 5910
+    },
+    {
+      "epoch": 3.869281045751634,
+      "grad_norm": 0.960086464881897,
+      "learning_rate": 0.0002,
+      "loss": 1.3576,
+      "step": 5920
+    },
+    {
+      "epoch": 3.8758169934640523,
+      "grad_norm": 0.9952049851417542,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 5930
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 0.7884191274642944,
+      "learning_rate": 0.0002,
+      "loss": 1.2027,
+      "step": 5940
+    },
+    {
+      "epoch": 3.888888888888889,
+      "grad_norm": 0.7461766600608826,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5950
+    },
+    {
+      "epoch": 3.895424836601307,
+      "grad_norm": 0.9594355821609497,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 5960
+    },
+    {
+      "epoch": 3.9019607843137254,
+      "grad_norm": 0.8179471492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.1164,
+      "step": 5970
+    },
+    {
+      "epoch": 3.9084967320261437,
+      "grad_norm": 0.8240267634391785,
+      "learning_rate": 0.0002,
+      "loss": 1.2421,
+      "step": 5980
+    },
+    {
+      "epoch": 3.915032679738562,
+      "grad_norm": 0.7462618350982666,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 5990
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 0.711207389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.2124,
+      "step": 6000
+    },
+    {
+      "epoch": 3.928104575163399,
+      "grad_norm": 0.6910956501960754,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 6010
+    },
+    {
+      "epoch": 3.9346405228758172,
+      "grad_norm": 0.749093770980835,
+      "learning_rate": 0.0002,
+      "loss": 1.2127,
+      "step": 6020
+    },
+    {
+      "epoch": 3.9411764705882355,
+      "grad_norm": 1.3332762718200684,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6030
+    },
+    {
+      "epoch": 3.947712418300654,
+      "grad_norm": 0.71457439661026,
+      "learning_rate": 0.0002,
+      "loss": 1.1442,
+      "step": 6040
+    },
+    {
+      "epoch": 3.954248366013072,
+      "grad_norm": 1.1205238103866577,
+      "learning_rate": 0.0002,
+      "loss": 1.339,
+      "step": 6050
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 0.6958928108215332,
+      "learning_rate": 0.0002,
+      "loss": 1.2962,
+      "step": 6060
+    },
+    {
+      "epoch": 3.9673202614379086,
+      "grad_norm": 0.7518056035041809,
+      "learning_rate": 0.0002,
+      "loss": 1.1802,
+      "step": 6070
+    },
+    {
+      "epoch": 3.973856209150327,
+      "grad_norm": 0.8010755777359009,
+      "learning_rate": 0.0002,
+      "loss": 1.1179,
+      "step": 6080
+    },
+    {
+      "epoch": 3.980392156862745,
+      "grad_norm": 0.7492658495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 6090
+    },
+    {
+      "epoch": 3.9869281045751634,
+      "grad_norm": 0.900704562664032,
+      "learning_rate": 0.0002,
+      "loss": 1.2113,
+      "step": 6100
+    },
+    {
+      "epoch": 3.9934640522875817,
+      "grad_norm": 0.7997331619262695,
+      "learning_rate": 0.0002,
+      "loss": 1.1106,
+      "step": 6110
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7163209319114685,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 6120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4113320112228394,
+      "eval_runtime": 33.7199,
+      "eval_samples_per_second": 12.93,
+      "eval_steps_per_second": 1.631,
+      "step": 6120
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 7.479861757083648e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..42757da2e98e15d76082d6943c5f99b90bf573a0
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae26adafc586de04059c8bcf1b7ea29a4bfc43656d2df3ccbc5fb98d67c85bd4
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7f0ba88abb7145c0f96e3d330e000a035284e4eb
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3adddafbad48fa0f66d361b088729b9594d7d901b40b54f0c49c037c3f9be4c
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bfd6bb6adb15181e548ba66c119648c547dc307a
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaf688fb3b028b2cbf957877240fe2f9bbb9f7be9a1796382a4722d9fb5b539f
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5612980a5f6d95b62c57b4b2e536723f259a0828
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b28dc3be2984476e1b98fe5a611b98bf32443bfdeadbedb0d6d08ca284c9087
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..47667316109e621fedc49ae6867683fcbf05d271
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/trainer_state.json
@@ -0,0 +1,5428 @@
+{
+  "best_metric": 1.4113320112228394,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120",
+  "epoch": 5.0,
+  "eval_steps": 10,
+  "global_step": 7650,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0065359477124183,
+      "grad_norm": 0.6926420331001282,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0130718954248366,
+      "grad_norm": 0.6656355857849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 4610
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 0.9901936650276184,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 4620
+    },
+    {
+      "epoch": 3.026143790849673,
+      "grad_norm": 0.6713474988937378,
+      "learning_rate": 0.0002,
+      "loss": 1.22,
+      "step": 4630
+    },
+    {
+      "epoch": 3.0326797385620914,
+      "grad_norm": 0.6199324131011963,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 4640
+    },
+    {
+      "epoch": 3.0392156862745097,
+      "grad_norm": 0.7180785536766052,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 4650
+    },
+    {
+      "epoch": 3.045751633986928,
+      "grad_norm": 0.8256588578224182,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 4660
+    },
+    {
+      "epoch": 3.052287581699346,
+      "grad_norm": 0.6637389063835144,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 4670
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 0.6980698108673096,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 4680
+    },
+    {
+      "epoch": 3.065359477124183,
+      "grad_norm": 0.8091534972190857,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 4690
+    },
+    {
+      "epoch": 3.0718954248366015,
+      "grad_norm": 0.5715174078941345,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 4700
+    },
+    {
+      "epoch": 3.0784313725490198,
+      "grad_norm": 0.735639750957489,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 4710
+    },
+    {
+      "epoch": 3.084967320261438,
+      "grad_norm": 0.7619708180427551,
+      "learning_rate": 0.0002,
+      "loss": 1.1522,
+      "step": 4720
+    },
+    {
+      "epoch": 3.0915032679738563,
+      "grad_norm": 1.263566017150879,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 4730
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 0.6600871682167053,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4740
+    },
+    {
+      "epoch": 3.104575163398693,
+      "grad_norm": 0.717792809009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 4750
+    },
+    {
+      "epoch": 3.111111111111111,
+      "grad_norm": 0.853714644908905,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1176470588235294,
+      "grad_norm": 1.1004153490066528,
+      "learning_rate": 0.0002,
+      "loss": 1.2031,
+      "step": 4770
+    },
+    {
+      "epoch": 3.1241830065359477,
+      "grad_norm": 0.8566235899925232,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 4780
+    },
+    {
+      "epoch": 3.130718954248366,
+      "grad_norm": 0.8315296173095703,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 4790
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 0.8020524978637695,
+      "learning_rate": 0.0002,
+      "loss": 1.32,
+      "step": 4800
+    },
+    {
+      "epoch": 3.1437908496732025,
+      "grad_norm": 0.7564275860786438,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 4810
+    },
+    {
+      "epoch": 3.150326797385621,
+      "grad_norm": 0.9077776670455933,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 4820
+    },
+    {
+      "epoch": 3.156862745098039,
+      "grad_norm": 0.6323099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 4830
+    },
+    {
+      "epoch": 3.1633986928104574,
+      "grad_norm": 0.6625368595123291,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 4840
+    },
+    {
+      "epoch": 3.1699346405228757,
+      "grad_norm": 0.8119261860847473,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 4850
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 0.6399450898170471,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 4860
+    },
+    {
+      "epoch": 3.183006535947712,
+      "grad_norm": 1.0659016370773315,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 4870
+    },
+    {
+      "epoch": 3.189542483660131,
+      "grad_norm": 0.8040369749069214,
+      "learning_rate": 0.0002,
+      "loss": 1.2914,
+      "step": 4880
+    },
+    {
+      "epoch": 3.196078431372549,
+      "grad_norm": 0.7784733176231384,
+      "learning_rate": 0.0002,
+      "loss": 1.1996,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2026143790849675,
+      "grad_norm": 0.9660294651985168,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 4900
+    },
+    {
+      "epoch": 3.2091503267973858,
+      "grad_norm": 1.0676977634429932,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 4910
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 0.5877565741539001,
+      "learning_rate": 0.0002,
+      "loss": 1.0083,
+      "step": 4920
+    },
+    {
+      "epoch": 3.2222222222222223,
+      "grad_norm": 0.6164032816886902,
+      "learning_rate": 0.0002,
+      "loss": 1.1046,
+      "step": 4930
+    },
+    {
+      "epoch": 3.2287581699346406,
+      "grad_norm": 0.7627606987953186,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 4940
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.7442803978919983,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 4950
+    },
+    {
+      "epoch": 3.241830065359477,
+      "grad_norm": 0.7277812361717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 4960
+    },
+    {
+      "epoch": 3.2483660130718954,
+      "grad_norm": 1.0301902294158936,
+      "learning_rate": 0.0002,
+      "loss": 1.2237,
+      "step": 4970
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 0.7798232436180115,
+      "learning_rate": 0.0002,
+      "loss": 1.1466,
+      "step": 4980
+    },
+    {
+      "epoch": 3.261437908496732,
+      "grad_norm": 1.210265874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.2142,
+      "step": 4990
+    },
+    {
+      "epoch": 3.2679738562091503,
+      "grad_norm": 0.6677713990211487,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2745098039215685,
+      "grad_norm": 1.0524500608444214,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 5010
+    },
+    {
+      "epoch": 3.281045751633987,
+      "grad_norm": 0.7091745734214783,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5020
+    },
+    {
+      "epoch": 3.287581699346405,
+      "grad_norm": 0.8523224592208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1891,
+      "step": 5030
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 0.6120608448982239,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3006535947712417,
+      "grad_norm": 0.7437472939491272,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 5050
+    },
+    {
+      "epoch": 3.30718954248366,
+      "grad_norm": 0.7611715197563171,
+      "learning_rate": 0.0002,
+      "loss": 1.1295,
+      "step": 5060
+    },
+    {
+      "epoch": 3.313725490196078,
+      "grad_norm": 0.7249704003334045,
+      "learning_rate": 0.0002,
+      "loss": 1.0531,
+      "step": 5070
+    },
+    {
+      "epoch": 3.3202614379084965,
+      "grad_norm": 0.7316247820854187,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5080
+    },
+    {
+      "epoch": 3.326797385620915,
+      "grad_norm": 0.562412440776825,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5090
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.7052176594734192,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5100
+    },
+    {
+      "epoch": 3.3398692810457518,
+      "grad_norm": 0.7714211344718933,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 5110
+    },
+    {
+      "epoch": 3.34640522875817,
+      "grad_norm": 1.0436055660247803,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5120
+    },
+    {
+      "epoch": 3.3529411764705883,
+      "grad_norm": 0.8867271542549133,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 5130
+    },
+    {
+      "epoch": 3.3594771241830066,
+      "grad_norm": 0.8371267914772034,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 5140
+    },
+    {
+      "epoch": 3.366013071895425,
+      "grad_norm": 0.7257837057113647,
+      "learning_rate": 0.0002,
+      "loss": 1.1073,
+      "step": 5150
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 0.7102002501487732,
+      "learning_rate": 0.0002,
+      "loss": 1.1162,
+      "step": 5160
+    },
+    {
+      "epoch": 3.3790849673202614,
+      "grad_norm": 0.7636350393295288,
+      "learning_rate": 0.0002,
+      "loss": 1.2056,
+      "step": 5170
+    },
+    {
+      "epoch": 3.3856209150326797,
+      "grad_norm": 0.6887359619140625,
+      "learning_rate": 0.0002,
+      "loss": 1.0708,
+      "step": 5180
+    },
+    {
+      "epoch": 3.392156862745098,
+      "grad_norm": 0.8141424655914307,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 5190
+    },
+    {
+      "epoch": 3.3986928104575163,
+      "grad_norm": 0.694423496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5200
+    },
+    {
+      "epoch": 3.4052287581699345,
+      "grad_norm": 0.914013683795929,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5210
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 0.8503239750862122,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 5220
+    },
+    {
+      "epoch": 3.418300653594771,
+      "grad_norm": 0.6196836233139038,
+      "learning_rate": 0.0002,
+      "loss": 1.2696,
+      "step": 5230
+    },
+    {
+      "epoch": 3.4248366013071894,
+      "grad_norm": 1.0760811567306519,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 5240
+    },
+    {
+      "epoch": 3.431372549019608,
+      "grad_norm": 0.6524698138237,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 5250
+    },
+    {
+      "epoch": 3.4379084967320264,
+      "grad_norm": 0.674467921257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5260
+    },
+    {
+      "epoch": 3.4444444444444446,
+      "grad_norm": 0.7690372467041016,
+      "learning_rate": 0.0002,
+      "loss": 1.1015,
+      "step": 5270
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 0.8751813769340515,
+      "learning_rate": 0.0002,
+      "loss": 1.2511,
+      "step": 5280
+    },
+    {
+      "epoch": 3.457516339869281,
+      "grad_norm": 0.750407874584198,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 5290
+    },
+    {
+      "epoch": 3.4640522875816995,
+      "grad_norm": 0.5991823077201843,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 5300
+    },
+    {
+      "epoch": 3.4705882352941178,
+      "grad_norm": 1.0164772272109985,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 5310
+    },
+    {
+      "epoch": 3.477124183006536,
+      "grad_norm": 0.8704105019569397,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 5320
+    },
+    {
+      "epoch": 3.4836601307189543,
+      "grad_norm": 0.709102213382721,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 5330
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 0.6273632049560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 5340
+    },
+    {
+      "epoch": 3.496732026143791,
+      "grad_norm": 0.6807359457015991,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 5350
+    },
+    {
+      "epoch": 3.503267973856209,
+      "grad_norm": 0.7085188627243042,
+      "learning_rate": 0.0002,
+      "loss": 1.131,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5098039215686274,
+      "grad_norm": 0.6938307881355286,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 5370
+    },
+    {
+      "epoch": 3.5163398692810457,
+      "grad_norm": 0.8544146418571472,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 5380
+    },
+    {
+      "epoch": 3.522875816993464,
+      "grad_norm": 0.7889642119407654,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 5390
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.7858421206474304,
+      "learning_rate": 0.0002,
+      "loss": 1.1691,
+      "step": 5400
+    },
+    {
+      "epoch": 3.5359477124183005,
+      "grad_norm": 0.8547123074531555,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5410
+    },
+    {
+      "epoch": 3.542483660130719,
+      "grad_norm": 0.8218181133270264,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 5420
+    },
+    {
+      "epoch": 3.549019607843137,
+      "grad_norm": 1.153623342514038,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 5430
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.1321099996566772,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 5440
+    },
+    {
+      "epoch": 3.5620915032679736,
+      "grad_norm": 0.9495334029197693,
+      "learning_rate": 0.0002,
+      "loss": 1.2224,
+      "step": 5450
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 0.8743821978569031,
+      "learning_rate": 0.0002,
+      "loss": 1.2869,
+      "step": 5460
+    },
+    {
+      "epoch": 3.57516339869281,
+      "grad_norm": 0.7513086795806885,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 5470
+    },
+    {
+      "epoch": 3.581699346405229,
+      "grad_norm": 1.0139480829238892,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 5480
+    },
+    {
+      "epoch": 3.588235294117647,
+      "grad_norm": 0.6615135073661804,
+      "learning_rate": 0.0002,
+      "loss": 1.1706,
+      "step": 5490
+    },
+    {
+      "epoch": 3.5947712418300655,
+      "grad_norm": 1.180798888206482,
+      "learning_rate": 0.0002,
+      "loss": 1.3906,
+      "step": 5500
+    },
+    {
+      "epoch": 3.6013071895424837,
+      "grad_norm": 0.7085279226303101,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 5510
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 0.540268063545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1623,
+      "step": 5520
+    },
+    {
+      "epoch": 3.6143790849673203,
+      "grad_norm": 0.7905671000480652,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 5530
+    },
+    {
+      "epoch": 3.6209150326797386,
+      "grad_norm": 0.8457717299461365,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 5540
+    },
+    {
+      "epoch": 3.627450980392157,
+      "grad_norm": 0.7102677822113037,
+      "learning_rate": 0.0002,
+      "loss": 1.1799,
+      "step": 5550
+    },
+    {
+      "epoch": 3.633986928104575,
+      "grad_norm": 0.7179514765739441,
+      "learning_rate": 0.0002,
+      "loss": 1.2394,
+      "step": 5560
+    },
+    {
+      "epoch": 3.6405228758169934,
+      "grad_norm": 1.0854148864746094,
+      "learning_rate": 0.0002,
+      "loss": 1.2019,
+      "step": 5570
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 0.8209951519966125,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5580
+    },
+    {
+      "epoch": 3.65359477124183,
+      "grad_norm": 0.6944138407707214,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 5590
+    },
+    {
+      "epoch": 3.6601307189542482,
+      "grad_norm": 0.7675473093986511,
+      "learning_rate": 0.0002,
+      "loss": 1.3226,
+      "step": 5600
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6683364510536194,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 5610
+    },
+    {
+      "epoch": 3.6732026143790852,
+      "grad_norm": 0.7920727133750916,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 5620
+    },
+    {
+      "epoch": 3.6797385620915035,
+      "grad_norm": 0.9440218806266785,
+      "learning_rate": 0.0002,
+      "loss": 1.2287,
+      "step": 5630
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 0.6600824594497681,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 5640
+    },
+    {
+      "epoch": 3.69281045751634,
+      "grad_norm": 0.6860619187355042,
+      "learning_rate": 0.0002,
+      "loss": 1.191,
+      "step": 5650
+    },
+    {
+      "epoch": 3.6993464052287583,
+      "grad_norm": 0.6579713225364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 5660
+    },
+    {
+      "epoch": 3.7058823529411766,
+      "grad_norm": 0.661081075668335,
+      "learning_rate": 0.0002,
+      "loss": 1.1464,
+      "step": 5670
+    },
+    {
+      "epoch": 3.712418300653595,
+      "grad_norm": 1.0968825817108154,
+      "learning_rate": 0.0002,
+      "loss": 1.289,
+      "step": 5680
+    },
+    {
+      "epoch": 3.718954248366013,
+      "grad_norm": 0.8066844940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.192,
+      "step": 5690
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 0.8341682553291321,
+      "learning_rate": 0.0002,
+      "loss": 1.2322,
+      "step": 5700
+    },
+    {
+      "epoch": 3.7320261437908497,
+      "grad_norm": 0.6682852506637573,
+      "learning_rate": 0.0002,
+      "loss": 1.1473,
+      "step": 5710
+    },
+    {
+      "epoch": 3.738562091503268,
+      "grad_norm": 0.898595929145813,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 5720
+    },
+    {
+      "epoch": 3.7450980392156863,
+      "grad_norm": 0.6876054406166077,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 5730
+    },
+    {
+      "epoch": 3.7516339869281046,
+      "grad_norm": 0.7817103266716003,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 5740
+    },
+    {
+      "epoch": 3.758169934640523,
+      "grad_norm": 0.5840168595314026,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 5750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 0.6263918876647949,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5760
+    },
+    {
+      "epoch": 3.7712418300653594,
+      "grad_norm": 0.7948952317237854,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 5770
+    },
+    {
+      "epoch": 3.7777777777777777,
+      "grad_norm": 0.6700998544692993,
+      "learning_rate": 0.0002,
+      "loss": 1.149,
+      "step": 5780
+    },
+    {
+      "epoch": 3.784313725490196,
+      "grad_norm": 1.1169519424438477,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 5790
+    },
+    {
+      "epoch": 3.7908496732026142,
+      "grad_norm": 0.8354471325874329,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 5800
+    },
+    {
+      "epoch": 3.7973856209150325,
+      "grad_norm": 0.6304181814193726,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 5810
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 0.6919655799865723,
+      "learning_rate": 0.0002,
+      "loss": 1.2059,
+      "step": 5820
+    },
+    {
+      "epoch": 3.810457516339869,
+      "grad_norm": 0.600385844707489,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5830
+    },
+    {
+      "epoch": 3.8169934640522873,
+      "grad_norm": 0.8406319618225098,
+      "learning_rate": 0.0002,
+      "loss": 1.2324,
+      "step": 5840
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.7594282031059265,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5850
+    },
+    {
+      "epoch": 3.8300653594771243,
+      "grad_norm": 0.8179879784584045,
+      "learning_rate": 0.0002,
+      "loss": 1.1903,
+      "step": 5860
+    },
+    {
+      "epoch": 3.8366013071895426,
+      "grad_norm": 1.141430377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 5870
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 0.6595550775527954,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 5880
+    },
+    {
+      "epoch": 3.849673202614379,
+      "grad_norm": 0.7499435544013977,
+      "learning_rate": 0.0002,
+      "loss": 1.2378,
+      "step": 5890
+    },
+    {
+      "epoch": 3.8562091503267975,
+      "grad_norm": 0.7851517200469971,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5900
+    },
+    {
+      "epoch": 3.8627450980392157,
+      "grad_norm": 1.0533545017242432,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 5910
+    },
+    {
+      "epoch": 3.869281045751634,
+      "grad_norm": 0.960086464881897,
+      "learning_rate": 0.0002,
+      "loss": 1.3576,
+      "step": 5920
+    },
+    {
+      "epoch": 3.8758169934640523,
+      "grad_norm": 0.9952049851417542,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 5930
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 0.7884191274642944,
+      "learning_rate": 0.0002,
+      "loss": 1.2027,
+      "step": 5940
+    },
+    {
+      "epoch": 3.888888888888889,
+      "grad_norm": 0.7461766600608826,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5950
+    },
+    {
+      "epoch": 3.895424836601307,
+      "grad_norm": 0.9594355821609497,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 5960
+    },
+    {
+      "epoch": 3.9019607843137254,
+      "grad_norm": 0.8179471492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.1164,
+      "step": 5970
+    },
+    {
+      "epoch": 3.9084967320261437,
+      "grad_norm": 0.8240267634391785,
+      "learning_rate": 0.0002,
+      "loss": 1.2421,
+      "step": 5980
+    },
+    {
+      "epoch": 3.915032679738562,
+      "grad_norm": 0.7462618350982666,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 5990
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 0.711207389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.2124,
+      "step": 6000
+    },
+    {
+      "epoch": 3.928104575163399,
+      "grad_norm": 0.6910956501960754,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 6010
+    },
+    {
+      "epoch": 3.9346405228758172,
+      "grad_norm": 0.749093770980835,
+      "learning_rate": 0.0002,
+      "loss": 1.2127,
+      "step": 6020
+    },
+    {
+      "epoch": 3.9411764705882355,
+      "grad_norm": 1.3332762718200684,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6030
+    },
+    {
+      "epoch": 3.947712418300654,
+      "grad_norm": 0.71457439661026,
+      "learning_rate": 0.0002,
+      "loss": 1.1442,
+      "step": 6040
+    },
+    {
+      "epoch": 3.954248366013072,
+      "grad_norm": 1.1205238103866577,
+      "learning_rate": 0.0002,
+      "loss": 1.339,
+      "step": 6050
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 0.6958928108215332,
+      "learning_rate": 0.0002,
+      "loss": 1.2962,
+      "step": 6060
+    },
+    {
+      "epoch": 3.9673202614379086,
+      "grad_norm": 0.7518056035041809,
+      "learning_rate": 0.0002,
+      "loss": 1.1802,
+      "step": 6070
+    },
+    {
+      "epoch": 3.973856209150327,
+      "grad_norm": 0.8010755777359009,
+      "learning_rate": 0.0002,
+      "loss": 1.1179,
+      "step": 6080
+    },
+    {
+      "epoch": 3.980392156862745,
+      "grad_norm": 0.7492658495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 6090
+    },
+    {
+      "epoch": 3.9869281045751634,
+      "grad_norm": 0.900704562664032,
+      "learning_rate": 0.0002,
+      "loss": 1.2113,
+      "step": 6100
+    },
+    {
+      "epoch": 3.9934640522875817,
+      "grad_norm": 0.7997331619262695,
+      "learning_rate": 0.0002,
+      "loss": 1.1106,
+      "step": 6110
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7163209319114685,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 6120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4113320112228394,
+      "eval_runtime": 33.7199,
+      "eval_samples_per_second": 12.93,
+      "eval_steps_per_second": 1.631,
+      "step": 6120
+    },
+    {
+      "epoch": 4.006535947712418,
+      "grad_norm": 0.9527022838592529,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 6130
+    },
+    {
+      "epoch": 4.0130718954248366,
+      "grad_norm": 0.7603210210800171,
+      "learning_rate": 0.0002,
+      "loss": 1.101,
+      "step": 6140
+    },
+    {
+      "epoch": 4.019607843137255,
+      "grad_norm": 1.127387523651123,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 6150
+    },
+    {
+      "epoch": 4.026143790849673,
+      "grad_norm": 0.8290133476257324,
+      "learning_rate": 0.0002,
+      "loss": 1.0734,
+      "step": 6160
+    },
+    {
+      "epoch": 4.032679738562091,
+      "grad_norm": 0.9912241101264954,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 6170
+    },
+    {
+      "epoch": 4.03921568627451,
+      "grad_norm": 0.947005033493042,
+      "learning_rate": 0.0002,
+      "loss": 1.0719,
+      "step": 6180
+    },
+    {
+      "epoch": 4.045751633986928,
+      "grad_norm": 0.707466185092926,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 6190
+    },
+    {
+      "epoch": 4.052287581699346,
+      "grad_norm": 1.0604327917099,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6200
+    },
+    {
+      "epoch": 4.0588235294117645,
+      "grad_norm": 0.7848685383796692,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6210
+    },
+    {
+      "epoch": 4.065359477124183,
+      "grad_norm": 0.8475256562232971,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 6220
+    },
+    {
+      "epoch": 4.071895424836601,
+      "grad_norm": 0.9759448766708374,
+      "learning_rate": 0.0002,
+      "loss": 1.1104,
+      "step": 6230
+    },
+    {
+      "epoch": 4.078431372549019,
+      "grad_norm": 0.9324519038200378,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 6240
+    },
+    {
+      "epoch": 4.084967320261438,
+      "grad_norm": 0.8723901510238647,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 6250
+    },
+    {
+      "epoch": 4.091503267973856,
+      "grad_norm": 0.8343415856361389,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6260
+    },
+    {
+      "epoch": 4.098039215686274,
+      "grad_norm": 0.7490310072898865,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 6270
+    },
+    {
+      "epoch": 4.104575163398692,
+      "grad_norm": 0.8961182832717896,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 6280
+    },
+    {
+      "epoch": 4.111111111111111,
+      "grad_norm": 0.7124854922294617,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 6290
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 0.8338138461112976,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 6300
+    },
+    {
+      "epoch": 4.124183006535947,
+      "grad_norm": 0.8075833320617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 6310
+    },
+    {
+      "epoch": 4.130718954248366,
+      "grad_norm": 0.8069391846656799,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 6320
+    },
+    {
+      "epoch": 4.137254901960785,
+      "grad_norm": 0.9567893147468567,
+      "learning_rate": 0.0002,
+      "loss": 0.948,
+      "step": 6330
+    },
+    {
+      "epoch": 4.143790849673203,
+      "grad_norm": 1.2184662818908691,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 6340
+    },
+    {
+      "epoch": 4.150326797385621,
+      "grad_norm": 1.030976414680481,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 6350
+    },
+    {
+      "epoch": 4.1568627450980395,
+      "grad_norm": 0.9749957323074341,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6360
+    },
+    {
+      "epoch": 4.163398692810458,
+      "grad_norm": 0.7089483141899109,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 6370
+    },
+    {
+      "epoch": 4.169934640522876,
+      "grad_norm": 1.1084946393966675,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 6380
+    },
+    {
+      "epoch": 4.176470588235294,
+      "grad_norm": 0.7998497486114502,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 6390
+    },
+    {
+      "epoch": 4.183006535947713,
+      "grad_norm": 0.8997811675071716,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 6400
+    },
+    {
+      "epoch": 4.189542483660131,
+      "grad_norm": 0.8359479904174805,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 6410
+    },
+    {
+      "epoch": 4.196078431372549,
+      "grad_norm": 0.9087472558021545,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 6420
+    },
+    {
+      "epoch": 4.2026143790849675,
+      "grad_norm": 1.1100451946258545,
+      "learning_rate": 0.0002,
+      "loss": 1.0657,
+      "step": 6430
+    },
+    {
+      "epoch": 4.209150326797386,
+      "grad_norm": 0.9376999735832214,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 6440
+    },
+    {
+      "epoch": 4.215686274509804,
+      "grad_norm": 0.8179266452789307,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 6450
+    },
+    {
+      "epoch": 4.222222222222222,
+      "grad_norm": 0.9953271746635437,
+      "learning_rate": 0.0002,
+      "loss": 1.0679,
+      "step": 6460
+    },
+    {
+      "epoch": 4.228758169934641,
+      "grad_norm": 0.8476650714874268,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 6470
+    },
+    {
+      "epoch": 4.235294117647059,
+      "grad_norm": 0.8406323194503784,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 6480
+    },
+    {
+      "epoch": 4.241830065359477,
+      "grad_norm": 0.819134533405304,
+      "learning_rate": 0.0002,
+      "loss": 1.057,
+      "step": 6490
+    },
+    {
+      "epoch": 4.248366013071895,
+      "grad_norm": 0.7764983773231506,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 6500
+    },
+    {
+      "epoch": 4.254901960784314,
+      "grad_norm": 0.8252112865447998,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 6510
+    },
+    {
+      "epoch": 4.261437908496732,
+      "grad_norm": 0.7941019535064697,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 6520
+    },
+    {
+      "epoch": 4.26797385620915,
+      "grad_norm": 0.7673905491828918,
+      "learning_rate": 0.0002,
+      "loss": 1.0296,
+      "step": 6530
+    },
+    {
+      "epoch": 4.2745098039215685,
+      "grad_norm": 0.8749890327453613,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6540
+    },
+    {
+      "epoch": 4.281045751633987,
+      "grad_norm": 0.7343207597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 6550
+    },
+    {
+      "epoch": 4.287581699346405,
+      "grad_norm": 1.2786651849746704,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6560
+    },
+    {
+      "epoch": 4.294117647058823,
+      "grad_norm": 1.316875696182251,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6570
+    },
+    {
+      "epoch": 4.300653594771242,
+      "grad_norm": 0.8349189162254333,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 6580
+    },
+    {
+      "epoch": 4.30718954248366,
+      "grad_norm": 0.7510647177696228,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6590
+    },
+    {
+      "epoch": 4.313725490196078,
+      "grad_norm": 0.932420551776886,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6600
+    },
+    {
+      "epoch": 4.3202614379084965,
+      "grad_norm": 0.8510616421699524,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 6610
+    },
+    {
+      "epoch": 4.326797385620915,
+      "grad_norm": 0.7661547064781189,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6620
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 1.0370930433273315,
+      "learning_rate": 0.0002,
+      "loss": 1.2064,
+      "step": 6630
+    },
+    {
+      "epoch": 4.339869281045751,
+      "grad_norm": 0.9302158951759338,
+      "learning_rate": 0.0002,
+      "loss": 1.1064,
+      "step": 6640
+    },
+    {
+      "epoch": 4.34640522875817,
+      "grad_norm": 0.9203811883926392,
+      "learning_rate": 0.0002,
+      "loss": 0.968,
+      "step": 6650
+    },
+    {
+      "epoch": 4.352941176470588,
+      "grad_norm": 0.9986332654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 6660
+    },
+    {
+      "epoch": 4.359477124183006,
+      "grad_norm": 0.8001713156700134,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6670
+    },
+    {
+      "epoch": 4.366013071895424,
+      "grad_norm": 0.829714298248291,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 6680
+    },
+    {
+      "epoch": 4.372549019607844,
+      "grad_norm": 0.8253079056739807,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 6690
+    },
+    {
+      "epoch": 4.379084967320262,
+      "grad_norm": 0.824666440486908,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 6700
+    },
+    {
+      "epoch": 4.38562091503268,
+      "grad_norm": 0.8872972130775452,
+      "learning_rate": 0.0002,
+      "loss": 1.1968,
+      "step": 6710
+    },
+    {
+      "epoch": 4.392156862745098,
+      "grad_norm": 0.8729761838912964,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 6720
+    },
+    {
+      "epoch": 4.398692810457517,
+      "grad_norm": 1.1367264986038208,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 6730
+    },
+    {
+      "epoch": 4.405228758169935,
+      "grad_norm": 0.9699058532714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0184,
+      "step": 6740
+    },
+    {
+      "epoch": 4.411764705882353,
+      "grad_norm": 0.8266763687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 6750
+    },
+    {
+      "epoch": 4.4183006535947715,
+      "grad_norm": 1.0249767303466797,
+      "learning_rate": 0.0002,
+      "loss": 1.0735,
+      "step": 6760
+    },
+    {
+      "epoch": 4.42483660130719,
+      "grad_norm": 0.73606938123703,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 6770
+    },
+    {
+      "epoch": 4.431372549019608,
+      "grad_norm": 1.4050679206848145,
+      "learning_rate": 0.0002,
+      "loss": 1.1037,
+      "step": 6780
+    },
+    {
+      "epoch": 4.437908496732026,
+      "grad_norm": 1.1114081144332886,
+      "learning_rate": 0.0002,
+      "loss": 1.1418,
+      "step": 6790
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8031067848205566,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 6800
+    },
+    {
+      "epoch": 4.450980392156863,
+      "grad_norm": 0.8513566851615906,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6810
+    },
+    {
+      "epoch": 4.457516339869281,
+      "grad_norm": 1.332741379737854,
+      "learning_rate": 0.0002,
+      "loss": 1.1852,
+      "step": 6820
+    },
+    {
+      "epoch": 4.4640522875816995,
+      "grad_norm": 1.5032578706741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 6830
+    },
+    {
+      "epoch": 4.470588235294118,
+      "grad_norm": 0.7677283883094788,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6840
+    },
+    {
+      "epoch": 4.477124183006536,
+      "grad_norm": 0.989148736000061,
+      "learning_rate": 0.0002,
+      "loss": 1.1501,
+      "step": 6850
+    },
+    {
+      "epoch": 4.483660130718954,
+      "grad_norm": 1.5316275358200073,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 6860
+    },
+    {
+      "epoch": 4.490196078431373,
+      "grad_norm": 0.9427124261856079,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 6870
+    },
+    {
+      "epoch": 4.496732026143791,
+      "grad_norm": 1.215287685394287,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 6880
+    },
+    {
+      "epoch": 4.503267973856209,
+      "grad_norm": 0.7286760210990906,
+      "learning_rate": 0.0002,
+      "loss": 1.0809,
+      "step": 6890
+    },
+    {
+      "epoch": 4.509803921568627,
+      "grad_norm": 0.874829888343811,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6900
+    },
+    {
+      "epoch": 4.516339869281046,
+      "grad_norm": 0.8058359622955322,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 6910
+    },
+    {
+      "epoch": 4.522875816993464,
+      "grad_norm": 1.248195767402649,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 6920
+    },
+    {
+      "epoch": 4.529411764705882,
+      "grad_norm": 0.8033645749092102,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 6930
+    },
+    {
+      "epoch": 4.5359477124183005,
+      "grad_norm": 1.7361950874328613,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 6940
+    },
+    {
+      "epoch": 4.542483660130719,
+      "grad_norm": 0.8058095574378967,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 6950
+    },
+    {
+      "epoch": 4.549019607843137,
+      "grad_norm": 1.254089593887329,
+      "learning_rate": 0.0002,
+      "loss": 1.0057,
+      "step": 6960
+    },
+    {
+      "epoch": 4.555555555555555,
+      "grad_norm": 0.9180455803871155,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 6970
+    },
+    {
+      "epoch": 4.562091503267974,
+      "grad_norm": 0.6677682399749756,
+      "learning_rate": 0.0002,
+      "loss": 1.0559,
+      "step": 6980
+    },
+    {
+      "epoch": 4.568627450980392,
+      "grad_norm": 0.8127354383468628,
+      "learning_rate": 0.0002,
+      "loss": 1.0453,
+      "step": 6990
+    },
+    {
+      "epoch": 4.57516339869281,
+      "grad_norm": 1.0263001918792725,
+      "learning_rate": 0.0002,
+      "loss": 1.0828,
+      "step": 7000
+    },
+    {
+      "epoch": 4.5816993464052285,
+      "grad_norm": 0.9641909003257751,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 7010
+    },
+    {
+      "epoch": 4.588235294117647,
+      "grad_norm": 0.9440861344337463,
+      "learning_rate": 0.0002,
+      "loss": 1.179,
+      "step": 7020
+    },
+    {
+      "epoch": 4.594771241830065,
+      "grad_norm": 0.9539011716842651,
+      "learning_rate": 0.0002,
+      "loss": 1.0931,
+      "step": 7030
+    },
+    {
+      "epoch": 4.601307189542483,
+      "grad_norm": 1.0449910163879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7040
+    },
+    {
+      "epoch": 4.607843137254902,
+      "grad_norm": 0.8766893744468689,
+      "learning_rate": 0.0002,
+      "loss": 0.9944,
+      "step": 7050
+    },
+    {
+      "epoch": 4.61437908496732,
+      "grad_norm": 0.6983462572097778,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 7060
+    },
+    {
+      "epoch": 4.620915032679738,
+      "grad_norm": 0.9505505561828613,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 7070
+    },
+    {
+      "epoch": 4.627450980392156,
+      "grad_norm": 1.2506657838821411,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7080
+    },
+    {
+      "epoch": 4.633986928104575,
+      "grad_norm": 0.9602801203727722,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 7090
+    },
+    {
+      "epoch": 4.640522875816993,
+      "grad_norm": 0.7398977875709534,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 7100
+    },
+    {
+      "epoch": 4.647058823529412,
+      "grad_norm": 1.3862425088882446,
+      "learning_rate": 0.0002,
+      "loss": 1.0769,
+      "step": 7110
+    },
+    {
+      "epoch": 4.65359477124183,
+      "grad_norm": 1.1451990604400635,
+      "learning_rate": 0.0002,
+      "loss": 1.0571,
+      "step": 7120
+    },
+    {
+      "epoch": 4.660130718954249,
+      "grad_norm": 0.9010422229766846,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 7130
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.7102518081665039,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 7140
+    },
+    {
+      "epoch": 4.673202614379085,
+      "grad_norm": 0.7963796257972717,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7150
+    },
+    {
+      "epoch": 4.6797385620915035,
+      "grad_norm": 0.7726007699966431,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 7160
+    },
+    {
+      "epoch": 4.686274509803922,
+      "grad_norm": 0.8097564578056335,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 7170
+    },
+    {
+      "epoch": 4.69281045751634,
+      "grad_norm": 0.9070925116539001,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 7180
+    },
+    {
+      "epoch": 4.699346405228758,
+      "grad_norm": 0.7543528079986572,
+      "learning_rate": 0.0002,
+      "loss": 1.0315,
+      "step": 7190
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.9900904893875122,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 7200
+    },
+    {
+      "epoch": 4.712418300653595,
+      "grad_norm": 0.8033412098884583,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 7210
+    },
+    {
+      "epoch": 4.718954248366013,
+      "grad_norm": 0.8440839052200317,
+      "learning_rate": 0.0002,
+      "loss": 1.1773,
+      "step": 7220
+    },
+    {
+      "epoch": 4.7254901960784315,
+      "grad_norm": 0.9325555562973022,
+      "learning_rate": 0.0002,
+      "loss": 1.1258,
+      "step": 7230
+    },
+    {
+      "epoch": 4.73202614379085,
+      "grad_norm": 0.7881146669387817,
+      "learning_rate": 0.0002,
+      "loss": 1.1384,
+      "step": 7240
+    },
+    {
+      "epoch": 4.738562091503268,
+      "grad_norm": 0.884453296661377,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7250
+    },
+    {
+      "epoch": 4.745098039215686,
+      "grad_norm": 0.9274539351463318,
+      "learning_rate": 0.0002,
+      "loss": 1.1036,
+      "step": 7260
+    },
+    {
+      "epoch": 4.751633986928105,
+      "grad_norm": 1.2367479801177979,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7270
+    },
+    {
+      "epoch": 4.758169934640523,
+      "grad_norm": 0.9499821066856384,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7280
+    },
+    {
+      "epoch": 4.764705882352941,
+      "grad_norm": 2.1918580532073975,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 7290
+    },
+    {
+      "epoch": 4.771241830065359,
+      "grad_norm": 0.8221880793571472,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 7300
+    },
+    {
+      "epoch": 4.777777777777778,
+      "grad_norm": 0.871972918510437,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7310
+    },
+    {
+      "epoch": 4.784313725490196,
+      "grad_norm": 0.8034510612487793,
+      "learning_rate": 0.0002,
+      "loss": 1.0599,
+      "step": 7320
+    },
+    {
+      "epoch": 4.790849673202614,
+      "grad_norm": 0.8959605693817139,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 7330
+    },
+    {
+      "epoch": 4.7973856209150325,
+      "grad_norm": 1.2326215505599976,
+      "learning_rate": 0.0002,
+      "loss": 1.0176,
+      "step": 7340
+    },
+    {
+      "epoch": 4.803921568627451,
+      "grad_norm": 0.9725791811943054,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7350
+    },
+    {
+      "epoch": 4.810457516339869,
+      "grad_norm": 0.7240816354751587,
+      "learning_rate": 0.0002,
+      "loss": 1.1229,
+      "step": 7360
+    },
+    {
+      "epoch": 4.816993464052287,
+      "grad_norm": 0.8265769481658936,
+      "learning_rate": 0.0002,
+      "loss": 1.0669,
+      "step": 7370
+    },
+    {
+      "epoch": 4.823529411764706,
+      "grad_norm": 0.8888696432113647,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 7380
+    },
+    {
+      "epoch": 4.830065359477124,
+      "grad_norm": 0.7776556015014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 7390
+    },
+    {
+      "epoch": 4.836601307189542,
+      "grad_norm": 0.8772371411323547,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7400
+    },
+    {
+      "epoch": 4.8431372549019605,
+      "grad_norm": 0.9786531925201416,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7410
+    },
+    {
+      "epoch": 4.849673202614379,
+      "grad_norm": 0.9059745073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7420
+    },
+    {
+      "epoch": 4.856209150326797,
+      "grad_norm": 0.7422552108764648,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 7430
+    },
+    {
+      "epoch": 4.862745098039216,
+      "grad_norm": 1.3040380477905273,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 7440
+    },
+    {
+      "epoch": 4.8692810457516345,
+      "grad_norm": 1.3278473615646362,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 7450
+    },
+    {
+      "epoch": 4.875816993464053,
+      "grad_norm": 1.2705849409103394,
+      "learning_rate": 0.0002,
+      "loss": 1.0713,
+      "step": 7460
+    },
+    {
+      "epoch": 4.882352941176471,
+      "grad_norm": 0.8837892413139343,
+      "learning_rate": 0.0002,
+      "loss": 1.0034,
+      "step": 7470
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 0.8670691251754761,
+      "learning_rate": 0.0002,
+      "loss": 1.1716,
+      "step": 7480
+    },
+    {
+      "epoch": 4.895424836601308,
+      "grad_norm": 0.9662758111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 7490
+    },
+    {
+      "epoch": 4.901960784313726,
+      "grad_norm": 0.8188302516937256,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 7500
+    },
+    {
+      "epoch": 4.908496732026144,
+      "grad_norm": 0.769442617893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 7510
+    },
+    {
+      "epoch": 4.915032679738562,
+      "grad_norm": 1.1465084552764893,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 7520
+    },
+    {
+      "epoch": 4.921568627450981,
+      "grad_norm": 1.253214955329895,
+      "learning_rate": 0.0002,
+      "loss": 1.0768,
+      "step": 7530
+    },
+    {
+      "epoch": 4.928104575163399,
+      "grad_norm": 0.7922375202178955,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 7540
+    },
+    {
+      "epoch": 4.934640522875817,
+      "grad_norm": 0.8306851387023926,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7550
+    },
+    {
+      "epoch": 4.9411764705882355,
+      "grad_norm": 0.8486151099205017,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 7560
+    },
+    {
+      "epoch": 4.947712418300654,
+      "grad_norm": 1.2601467370986938,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7570
+    },
+    {
+      "epoch": 4.954248366013072,
+      "grad_norm": 0.7980747818946838,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 7580
+    },
+    {
+      "epoch": 4.96078431372549,
+      "grad_norm": 0.8653254508972168,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 7590
+    },
+    {
+      "epoch": 4.967320261437909,
+      "grad_norm": 0.9680571556091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 7600
+    },
+    {
+      "epoch": 4.973856209150327,
+      "grad_norm": 0.9554466605186462,
+      "learning_rate": 0.0002,
+      "loss": 1.1795,
+      "step": 7610
+    },
+    {
+      "epoch": 4.980392156862745,
+      "grad_norm": 1.3693897724151611,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 7620
+    },
+    {
+      "epoch": 4.9869281045751634,
+      "grad_norm": 0.7809282541275024,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 7630
+    },
+    {
+      "epoch": 4.993464052287582,
+      "grad_norm": 0.7528006434440613,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 7640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7491309642791748,
+      "learning_rate": 0.0002,
+      "loss": 0.9951,
+      "step": 7650
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.4197258949279785,
+      "eval_runtime": 33.6327,
+      "eval_samples_per_second": 12.964,
+      "eval_steps_per_second": 1.635,
+      "step": 7650
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9.34982719635456e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-7650/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/README.md b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..830a14f7db2734beb59f320973504e45a3fe87f5
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..600a5ae79fa5bbcdea8bd42ae99abf77134a3287
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_model.safetensors b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0a815a13a8a67babedd2d8bbc4583115dc94bdcc
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fbd4b8470273ecd80cc3a19d933def3cdeffe6e1729bef1ba024865e4886a51
+size 29500848
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/optimizer.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1dc76144022dd2b16db1d8ed4a853c1f26910035
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80e58d2e4b8f746e1207bfe178bcfe44664c6c98e027ad47204c8b6738773847
+size 15064314
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/rng_state.pth b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..56eef58f5144cf90942b24fa10bf716b36463061
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78bd575ea1ef2db1c66988ade36263f1e07ea2252e737172bd8ae82ecf081c4a
+size 14244
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/scheduler.pt b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7e90f720eb8a510417707100c1ea1a627e13e392
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77fc51c9f2de817ea625082e092892004411da15fca41ad0dc86bc55cb1abd77
+size 1064
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/trainer_state.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..452e63cd783d5ca3a2a3136c5f42c705ab75dd78
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/trainer_state.json
@@ -0,0 +1,6507 @@
+{
+  "best_metric": 1.4113320112228394,
+  "best_model_checkpoint": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 9180,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006535947712418301,
+      "grad_norm": 1.5105072259902954,
+      "learning_rate": 0.0002,
+      "loss": 4.7451,
+      "step": 10
+    },
+    {
+      "epoch": 0.013071895424836602,
+      "grad_norm": 2.1156165599823,
+      "learning_rate": 0.0002,
+      "loss": 3.3158,
+      "step": 20
+    },
+    {
+      "epoch": 0.0196078431372549,
+      "grad_norm": 1.0578808784484863,
+      "learning_rate": 0.0002,
+      "loss": 2.643,
+      "step": 30
+    },
+    {
+      "epoch": 0.026143790849673203,
+      "grad_norm": 2.725064516067505,
+      "learning_rate": 0.0002,
+      "loss": 2.3948,
+      "step": 40
+    },
+    {
+      "epoch": 0.032679738562091505,
+      "grad_norm": 2.9575750827789307,
+      "learning_rate": 0.0002,
+      "loss": 2.3134,
+      "step": 50
+    },
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 1.2158117294311523,
+      "learning_rate": 0.0002,
+      "loss": 2.2778,
+      "step": 60
+    },
+    {
+      "epoch": 0.0457516339869281,
+      "grad_norm": 1.0850954055786133,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05228758169934641,
+      "grad_norm": 1.299196720123291,
+      "learning_rate": 0.0002,
+      "loss": 1.8872,
+      "step": 80
+    },
+    {
+      "epoch": 0.058823529411764705,
+      "grad_norm": 0.8310191035270691,
+      "learning_rate": 0.0002,
+      "loss": 1.947,
+      "step": 90
+    },
+    {
+      "epoch": 0.06535947712418301,
+      "grad_norm": 0.9854435920715332,
+      "learning_rate": 0.0002,
+      "loss": 1.9098,
+      "step": 100
+    },
+    {
+      "epoch": 0.0718954248366013,
+      "grad_norm": 0.7951157689094543,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 110
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 0.7593062520027161,
+      "learning_rate": 0.0002,
+      "loss": 1.9035,
+      "step": 120
+    },
+    {
+      "epoch": 0.08496732026143791,
+      "grad_norm": 0.6783032417297363,
+      "learning_rate": 0.0002,
+      "loss": 1.8517,
+      "step": 130
+    },
+    {
+      "epoch": 0.0915032679738562,
+      "grad_norm": 0.8350756764411926,
+      "learning_rate": 0.0002,
+      "loss": 1.6805,
+      "step": 140
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.0203173160552979,
+      "learning_rate": 0.0002,
+      "loss": 1.6123,
+      "step": 150
+    },
+    {
+      "epoch": 0.10457516339869281,
+      "grad_norm": 0.8820539712905884,
+      "learning_rate": 0.0002,
+      "loss": 1.7248,
+      "step": 160
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.7286128997802734,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 170
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 0.7874041795730591,
+      "learning_rate": 0.0002,
+      "loss": 1.8841,
+      "step": 180
+    },
+    {
+      "epoch": 0.12418300653594772,
+      "grad_norm": 0.6630475521087646,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 190
+    },
+    {
+      "epoch": 0.13071895424836602,
+      "grad_norm": 0.686413586139679,
+      "learning_rate": 0.0002,
+      "loss": 1.6149,
+      "step": 200
+    },
+    {
+      "epoch": 0.13725490196078433,
+      "grad_norm": 0.7793629765510559,
+      "learning_rate": 0.0002,
+      "loss": 1.6227,
+      "step": 210
+    },
+    {
+      "epoch": 0.1437908496732026,
+      "grad_norm": 0.6893141865730286,
+      "learning_rate": 0.0002,
+      "loss": 1.7223,
+      "step": 220
+    },
+    {
+      "epoch": 0.1503267973856209,
+      "grad_norm": 0.5804724097251892,
+      "learning_rate": 0.0002,
+      "loss": 1.6808,
+      "step": 230
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 0.6053574085235596,
+      "learning_rate": 0.0002,
+      "loss": 1.5578,
+      "step": 240
+    },
+    {
+      "epoch": 0.16339869281045752,
+      "grad_norm": 0.7566025853157043,
+      "learning_rate": 0.0002,
+      "loss": 1.7394,
+      "step": 250
+    },
+    {
+      "epoch": 0.16993464052287582,
+      "grad_norm": 0.6112990975379944,
+      "learning_rate": 0.0002,
+      "loss": 1.6216,
+      "step": 260
+    },
+    {
+      "epoch": 0.17647058823529413,
+      "grad_norm": 0.6839066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.5564,
+      "step": 270
+    },
+    {
+      "epoch": 0.1830065359477124,
+      "grad_norm": 0.6368117928504944,
+      "learning_rate": 0.0002,
+      "loss": 1.7129,
+      "step": 280
+    },
+    {
+      "epoch": 0.1895424836601307,
+      "grad_norm": 0.6144475936889648,
+      "learning_rate": 0.0002,
+      "loss": 1.5646,
+      "step": 290
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 0.6743767261505127,
+      "learning_rate": 0.0002,
+      "loss": 1.8383,
+      "step": 300
+    },
+    {
+      "epoch": 0.20261437908496732,
+      "grad_norm": 0.6807955503463745,
+      "learning_rate": 0.0002,
+      "loss": 1.421,
+      "step": 310
+    },
+    {
+      "epoch": 0.20915032679738563,
+      "grad_norm": 0.6717963814735413,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 320
+    },
+    {
+      "epoch": 0.21568627450980393,
+      "grad_norm": 0.5917780995368958,
+      "learning_rate": 0.0002,
+      "loss": 1.6842,
+      "step": 330
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.6783658862113953,
+      "learning_rate": 0.0002,
+      "loss": 1.6264,
+      "step": 340
+    },
+    {
+      "epoch": 0.22875816993464052,
+      "grad_norm": 0.5820256471633911,
+      "learning_rate": 0.0002,
+      "loss": 1.4635,
+      "step": 350
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 0.5345938801765442,
+      "learning_rate": 0.0002,
+      "loss": 1.6514,
+      "step": 360
+    },
+    {
+      "epoch": 0.24183006535947713,
+      "grad_norm": 0.755929172039032,
+      "learning_rate": 0.0002,
+      "loss": 1.6441,
+      "step": 370
+    },
+    {
+      "epoch": 0.24836601307189543,
+      "grad_norm": 0.6183189749717712,
+      "learning_rate": 0.0002,
+      "loss": 1.5177,
+      "step": 380
+    },
+    {
+      "epoch": 0.2549019607843137,
+      "grad_norm": 0.7277782559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 390
+    },
+    {
+      "epoch": 0.26143790849673204,
+      "grad_norm": 0.9998756051063538,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 400
+    },
+    {
+      "epoch": 0.2679738562091503,
+      "grad_norm": 0.7523853778839111,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 410
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 0.6548714637756348,
+      "learning_rate": 0.0002,
+      "loss": 1.5649,
+      "step": 420
+    },
+    {
+      "epoch": 0.28104575163398693,
+      "grad_norm": 0.6979796290397644,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 430
+    },
+    {
+      "epoch": 0.2875816993464052,
+      "grad_norm": 0.840915322303772,
+      "learning_rate": 0.0002,
+      "loss": 1.5927,
+      "step": 440
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.6142978072166443,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 450
+    },
+    {
+      "epoch": 0.3006535947712418,
+      "grad_norm": 0.9482691884040833,
+      "learning_rate": 0.0002,
+      "loss": 1.4903,
+      "step": 460
+    },
+    {
+      "epoch": 0.30718954248366015,
+      "grad_norm": 0.7001156806945801,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 470
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 0.6665455102920532,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 480
+    },
+    {
+      "epoch": 0.3202614379084967,
+      "grad_norm": 0.6012697815895081,
+      "learning_rate": 0.0002,
+      "loss": 1.587,
+      "step": 490
+    },
+    {
+      "epoch": 0.32679738562091504,
+      "grad_norm": 0.8770062327384949,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 500
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.7029962539672852,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 510
+    },
+    {
+      "epoch": 0.33986928104575165,
+      "grad_norm": 0.6682832837104797,
+      "learning_rate": 0.0002,
+      "loss": 1.4435,
+      "step": 520
+    },
+    {
+      "epoch": 0.3464052287581699,
+      "grad_norm": 0.5548969507217407,
+      "learning_rate": 0.0002,
+      "loss": 1.4242,
+      "step": 530
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 0.6640702486038208,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 540
+    },
+    {
+      "epoch": 0.35947712418300654,
+      "grad_norm": 0.656292200088501,
+      "learning_rate": 0.0002,
+      "loss": 1.4998,
+      "step": 550
+    },
+    {
+      "epoch": 0.3660130718954248,
+      "grad_norm": 0.618910551071167,
+      "learning_rate": 0.0002,
+      "loss": 1.5415,
+      "step": 560
+    },
+    {
+      "epoch": 0.37254901960784315,
+      "grad_norm": 0.644859790802002,
+      "learning_rate": 0.0002,
+      "loss": 1.5178,
+      "step": 570
+    },
+    {
+      "epoch": 0.3790849673202614,
+      "grad_norm": 0.679042398929596,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 580
+    },
+    {
+      "epoch": 0.38562091503267976,
+      "grad_norm": 0.980681836605072,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 590
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 0.632219672203064,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 600
+    },
+    {
+      "epoch": 0.39869281045751637,
+      "grad_norm": 0.7003744840621948,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 610
+    },
+    {
+      "epoch": 0.40522875816993464,
+      "grad_norm": 0.7090577483177185,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 620
+    },
+    {
+      "epoch": 0.4117647058823529,
+      "grad_norm": 0.657819926738739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 630
+    },
+    {
+      "epoch": 0.41830065359477125,
+      "grad_norm": 0.7034208178520203,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 640
+    },
+    {
+      "epoch": 0.42483660130718953,
+      "grad_norm": 0.7274866104125977,
+      "learning_rate": 0.0002,
+      "loss": 1.5357,
+      "step": 650
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 0.5876233577728271,
+      "learning_rate": 0.0002,
+      "loss": 1.6304,
+      "step": 660
+    },
+    {
+      "epoch": 0.43790849673202614,
+      "grad_norm": 0.595494270324707,
+      "learning_rate": 0.0002,
+      "loss": 1.7683,
+      "step": 670
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.8253804445266724,
+      "learning_rate": 0.0002,
+      "loss": 1.5117,
+      "step": 680
+    },
+    {
+      "epoch": 0.45098039215686275,
+      "grad_norm": 0.652225911617279,
+      "learning_rate": 0.0002,
+      "loss": 1.5199,
+      "step": 690
+    },
+    {
+      "epoch": 0.45751633986928103,
+      "grad_norm": 0.6242014169692993,
+      "learning_rate": 0.0002,
+      "loss": 1.5419,
+      "step": 700
+    },
+    {
+      "epoch": 0.46405228758169936,
+      "grad_norm": 0.7283986210823059,
+      "learning_rate": 0.0002,
+      "loss": 1.53,
+      "step": 710
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 0.7016081213951111,
+      "learning_rate": 0.0002,
+      "loss": 1.43,
+      "step": 720
+    },
+    {
+      "epoch": 0.477124183006536,
+      "grad_norm": 0.5211893916130066,
+      "learning_rate": 0.0002,
+      "loss": 1.4626,
+      "step": 730
+    },
+    {
+      "epoch": 0.48366013071895425,
+      "grad_norm": 0.6221150159835815,
+      "learning_rate": 0.0002,
+      "loss": 1.6885,
+      "step": 740
+    },
+    {
+      "epoch": 0.49019607843137253,
+      "grad_norm": 0.76594477891922,
+      "learning_rate": 0.0002,
+      "loss": 1.5677,
+      "step": 750
+    },
+    {
+      "epoch": 0.49673202614379086,
+      "grad_norm": 0.5777859091758728,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 760
+    },
+    {
+      "epoch": 0.5032679738562091,
+      "grad_norm": 0.5793519616127014,
+      "learning_rate": 0.0002,
+      "loss": 1.5253,
+      "step": 770
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 0.5425786375999451,
+      "learning_rate": 0.0002,
+      "loss": 1.3562,
+      "step": 780
+    },
+    {
+      "epoch": 0.5163398692810458,
+      "grad_norm": 0.6004197001457214,
+      "learning_rate": 0.0002,
+      "loss": 1.3398,
+      "step": 790
+    },
+    {
+      "epoch": 0.5228758169934641,
+      "grad_norm": 0.7167016863822937,
+      "learning_rate": 0.0002,
+      "loss": 1.5346,
+      "step": 800
+    },
+    {
+      "epoch": 0.5294117647058824,
+      "grad_norm": 0.710218071937561,
+      "learning_rate": 0.0002,
+      "loss": 1.48,
+      "step": 810
+    },
+    {
+      "epoch": 0.5359477124183006,
+      "grad_norm": 0.699528694152832,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 820
+    },
+    {
+      "epoch": 0.5424836601307189,
+      "grad_norm": 0.579629123210907,
+      "learning_rate": 0.0002,
+      "loss": 1.6014,
+      "step": 830
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 0.595407247543335,
+      "learning_rate": 0.0002,
+      "loss": 1.3894,
+      "step": 840
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.544563889503479,
+      "learning_rate": 0.0002,
+      "loss": 1.6394,
+      "step": 850
+    },
+    {
+      "epoch": 0.5620915032679739,
+      "grad_norm": 0.553166389465332,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 860
+    },
+    {
+      "epoch": 0.5686274509803921,
+      "grad_norm": 0.5645018815994263,
+      "learning_rate": 0.0002,
+      "loss": 1.5155,
+      "step": 870
+    },
+    {
+      "epoch": 0.5751633986928104,
+      "grad_norm": 0.6576932668685913,
+      "learning_rate": 0.0002,
+      "loss": 1.7019,
+      "step": 880
+    },
+    {
+      "epoch": 0.5816993464052288,
+      "grad_norm": 0.6684197187423706,
+      "learning_rate": 0.0002,
+      "loss": 1.5891,
+      "step": 890
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.6706975698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.5348,
+      "step": 900
+    },
+    {
+      "epoch": 0.5947712418300654,
+      "grad_norm": 0.6762327551841736,
+      "learning_rate": 0.0002,
+      "loss": 1.4038,
+      "step": 910
+    },
+    {
+      "epoch": 0.6013071895424836,
+      "grad_norm": 0.764032244682312,
+      "learning_rate": 0.0002,
+      "loss": 1.61,
+      "step": 920
+    },
+    {
+      "epoch": 0.6078431372549019,
+      "grad_norm": 0.6996400952339172,
+      "learning_rate": 0.0002,
+      "loss": 1.436,
+      "step": 930
+    },
+    {
+      "epoch": 0.6143790849673203,
+      "grad_norm": 0.686735987663269,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 940
+    },
+    {
+      "epoch": 0.6209150326797386,
+      "grad_norm": 0.6086131930351257,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 950
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 0.5627856850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.4457,
+      "step": 960
+    },
+    {
+      "epoch": 0.6339869281045751,
+      "grad_norm": 0.5781503319740295,
+      "learning_rate": 0.0002,
+      "loss": 1.506,
+      "step": 970
+    },
+    {
+      "epoch": 0.6405228758169934,
+      "grad_norm": 0.6347246766090393,
+      "learning_rate": 0.0002,
+      "loss": 1.5668,
+      "step": 980
+    },
+    {
+      "epoch": 0.6470588235294118,
+      "grad_norm": 0.6581300497055054,
+      "learning_rate": 0.0002,
+      "loss": 1.3819,
+      "step": 990
+    },
+    {
+      "epoch": 0.6535947712418301,
+      "grad_norm": 0.8343676924705505,
+      "learning_rate": 0.0002,
+      "loss": 1.6425,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6601307189542484,
+      "grad_norm": 0.5708910226821899,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.6832585334777832,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 1020
+    },
+    {
+      "epoch": 0.673202614379085,
+      "grad_norm": 0.5767837166786194,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6797385620915033,
+      "grad_norm": 0.5637745261192322,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6862745098039216,
+      "grad_norm": 0.8193050026893616,
+      "learning_rate": 0.0002,
+      "loss": 1.4325,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6928104575163399,
+      "grad_norm": 0.6157439351081848,
+      "learning_rate": 0.0002,
+      "loss": 1.4196,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6993464052287581,
+      "grad_norm": 0.7476664781570435,
+      "learning_rate": 0.0002,
+      "loss": 1.5547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 0.8569361567497253,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7124183006535948,
+      "grad_norm": 0.5671911835670471,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7189542483660131,
+      "grad_norm": 0.5151128768920898,
+      "learning_rate": 0.0002,
+      "loss": 1.5398,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7254901960784313,
+      "grad_norm": 0.568037211894989,
+      "learning_rate": 0.0002,
+      "loss": 1.4848,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7320261437908496,
+      "grad_norm": 0.6756396889686584,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 1120
+    },
+    {
+      "epoch": 0.738562091503268,
+      "grad_norm": 0.638975977897644,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 0.7103341221809387,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7516339869281046,
+      "grad_norm": 0.7403952479362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7581699346405228,
+      "grad_norm": 0.6266511082649231,
+      "learning_rate": 0.0002,
+      "loss": 1.4757,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7647058823529411,
+      "grad_norm": 0.5939070582389832,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7712418300653595,
+      "grad_norm": 0.5735430717468262,
+      "learning_rate": 0.0002,
+      "loss": 1.4145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.5155234932899475,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 0.5115423202514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4942,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7908496732026143,
+      "grad_norm": 0.693588137626648,
+      "learning_rate": 0.0002,
+      "loss": 1.4508,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7973856209150327,
+      "grad_norm": 0.5504693984985352,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 1220
+    },
+    {
+      "epoch": 0.803921568627451,
+      "grad_norm": 0.5555992126464844,
+      "learning_rate": 0.0002,
+      "loss": 1.5412,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8104575163398693,
+      "grad_norm": 0.7211785316467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5506,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8169934640522876,
+      "grad_norm": 0.735003650188446,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 0.5245152711868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5836,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8300653594771242,
+      "grad_norm": 0.5883445739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.4505,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8366013071895425,
+      "grad_norm": 0.6835859417915344,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8431372549019608,
+      "grad_norm": 0.6592142581939697,
+      "learning_rate": 0.0002,
+      "loss": 1.5526,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8496732026143791,
+      "grad_norm": 0.6087474226951599,
+      "learning_rate": 0.0002,
+      "loss": 1.52,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8562091503267973,
+      "grad_norm": 0.565387487411499,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 0.7363151907920837,
+      "learning_rate": 0.0002,
+      "loss": 1.4809,
+      "step": 1320
+    },
+    {
+      "epoch": 0.869281045751634,
+      "grad_norm": 0.5964524149894714,
+      "learning_rate": 0.0002,
+      "loss": 1.5683,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8758169934640523,
+      "grad_norm": 0.5169979929924011,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.7063422799110413,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 0.7261926531791687,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8954248366013072,
+      "grad_norm": 0.6759744882583618,
+      "learning_rate": 0.0002,
+      "loss": 1.3619,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 0.675051212310791,
+      "learning_rate": 0.0002,
+      "loss": 1.4079,
+      "step": 1380
+    },
+    {
+      "epoch": 0.9084967320261438,
+      "grad_norm": 0.5613595843315125,
+      "learning_rate": 0.0002,
+      "loss": 1.6606,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9150326797385621,
+      "grad_norm": 0.611732006072998,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9215686274509803,
+      "grad_norm": 0.6365187168121338,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9281045751633987,
+      "grad_norm": 0.7810426354408264,
+      "learning_rate": 0.0002,
+      "loss": 1.7832,
+      "step": 1420
+    },
+    {
+      "epoch": 0.934640522875817,
+      "grad_norm": 0.593891441822052,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 0.761585533618927,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9477124183006536,
+      "grad_norm": 0.6114464998245239,
+      "learning_rate": 0.0002,
+      "loss": 1.589,
+      "step": 1450
+    },
+    {
+      "epoch": 0.954248366013072,
+      "grad_norm": 0.601044774055481,
+      "learning_rate": 0.0002,
+      "loss": 1.4973,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9607843137254902,
+      "grad_norm": 0.5484876036643982,
+      "learning_rate": 0.0002,
+      "loss": 1.4162,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9673202614379085,
+      "grad_norm": 0.5383428335189819,
+      "learning_rate": 0.0002,
+      "loss": 1.4825,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9738562091503268,
+      "grad_norm": 0.648106575012207,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 0.6847249865531921,
+      "learning_rate": 0.0002,
+      "loss": 1.3638,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9869281045751634,
+      "grad_norm": 0.6361058354377747,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9934640522875817,
+      "grad_norm": 0.646392285823822,
+      "learning_rate": 0.0002,
+      "loss": 1.5131,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5391159057617188,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.4715123176574707,
+      "eval_runtime": 30.5701,
+      "eval_samples_per_second": 14.262,
+      "eval_steps_per_second": 1.799,
+      "step": 1530
+    },
+    {
+      "epoch": 1.0065359477124183,
+      "grad_norm": 0.5468988418579102,
+      "learning_rate": 0.0002,
+      "loss": 1.4827,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0130718954248366,
+      "grad_norm": 0.629940927028656,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 0.6411303281784058,
+      "learning_rate": 0.0002,
+      "loss": 1.4259,
+      "step": 1560
+    },
+    {
+      "epoch": 1.026143790849673,
+      "grad_norm": 0.5619024038314819,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0326797385620916,
+      "grad_norm": 0.6093462705612183,
+      "learning_rate": 0.0002,
+      "loss": 1.6086,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0392156862745099,
+      "grad_norm": 0.5543286204338074,
+      "learning_rate": 0.0002,
+      "loss": 1.4547,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0457516339869282,
+      "grad_norm": 0.6079006195068359,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0522875816993464,
+      "grad_norm": 0.6240813136100769,
+      "learning_rate": 0.0002,
+      "loss": 1.4574,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 0.6141977310180664,
+      "learning_rate": 0.0002,
+      "loss": 1.3504,
+      "step": 1620
+    },
+    {
+      "epoch": 1.065359477124183,
+      "grad_norm": 0.5920178294181824,
+      "learning_rate": 0.0002,
+      "loss": 1.3668,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0718954248366013,
+      "grad_norm": 0.47620782256126404,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0784313725490196,
+      "grad_norm": 0.6826292872428894,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0849673202614378,
+      "grad_norm": 0.6182006597518921,
+      "learning_rate": 0.0002,
+      "loss": 1.2285,
+      "step": 1660
+    },
+    {
+      "epoch": 1.091503267973856,
+      "grad_norm": 0.57639479637146,
+      "learning_rate": 0.0002,
+      "loss": 1.2907,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 0.6696860194206238,
+      "learning_rate": 0.0002,
+      "loss": 1.4575,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1045751633986929,
+      "grad_norm": 0.699221670627594,
+      "learning_rate": 0.0002,
+      "loss": 1.4104,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 0.7138059139251709,
+      "learning_rate": 0.0002,
+      "loss": 1.3667,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1176470588235294,
+      "grad_norm": 0.6930422186851501,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1241830065359477,
+      "grad_norm": 0.7484048008918762,
+      "learning_rate": 0.0002,
+      "loss": 1.5033,
+      "step": 1720
+    },
+    {
+      "epoch": 1.130718954248366,
+      "grad_norm": 0.5820090174674988,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 0.7143406867980957,
+      "learning_rate": 0.0002,
+      "loss": 1.3704,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1437908496732025,
+      "grad_norm": 0.5597584247589111,
+      "learning_rate": 0.0002,
+      "loss": 1.277,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1503267973856208,
+      "grad_norm": 0.5171173214912415,
+      "learning_rate": 0.0002,
+      "loss": 1.5403,
+      "step": 1760
+    },
+    {
+      "epoch": 1.156862745098039,
+      "grad_norm": 0.5951920747756958,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1633986928104576,
+      "grad_norm": 0.7506247758865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2929,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1699346405228759,
+      "grad_norm": 0.5936487913131714,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.688450038433075,
+      "learning_rate": 0.0002,
+      "loss": 1.3567,
+      "step": 1800
+    },
+    {
+      "epoch": 1.1830065359477124,
+      "grad_norm": 0.671623170375824,
+      "learning_rate": 0.0002,
+      "loss": 1.314,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1895424836601307,
+      "grad_norm": 0.6911860704421997,
+      "learning_rate": 0.0002,
+      "loss": 1.3803,
+      "step": 1820
+    },
+    {
+      "epoch": 1.196078431372549,
+      "grad_norm": 0.60726398229599,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2026143790849673,
+      "grad_norm": 0.7542088627815247,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2091503267973855,
+      "grad_norm": 0.6810969710350037,
+      "learning_rate": 0.0002,
+      "loss": 1.4343,
+      "step": 1850
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 0.579741895198822,
+      "learning_rate": 0.0002,
+      "loss": 1.446,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 0.9925695657730103,
+      "learning_rate": 0.0002,
+      "loss": 1.4564,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2287581699346406,
+      "grad_norm": 0.5919767618179321,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2352941176470589,
+      "grad_norm": 0.7377090454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5015,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2418300653594772,
+      "grad_norm": 0.5753688812255859,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2483660130718954,
+      "grad_norm": 0.6362486481666565,
+      "learning_rate": 0.0002,
+      "loss": 1.3543,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 0.5747467875480652,
+      "learning_rate": 0.0002,
+      "loss": 1.4153,
+      "step": 1920
+    },
+    {
+      "epoch": 1.261437908496732,
+      "grad_norm": 0.6831939220428467,
+      "learning_rate": 0.0002,
+      "loss": 1.5082,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2679738562091503,
+      "grad_norm": 0.6414040327072144,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2745098039215685,
+      "grad_norm": 0.5613330006599426,
+      "learning_rate": 0.0002,
+      "loss": 1.5099,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2810457516339868,
+      "grad_norm": 0.5838454961776733,
+      "learning_rate": 0.0002,
+      "loss": 1.377,
+      "step": 1960
+    },
+    {
+      "epoch": 1.287581699346405,
+      "grad_norm": 0.5367192029953003,
+      "learning_rate": 0.0002,
+      "loss": 1.3548,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 0.5829346776008606,
+      "learning_rate": 0.0002,
+      "loss": 1.4602,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3006535947712419,
+      "grad_norm": 0.756534218788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3821,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3071895424836601,
+      "grad_norm": 0.48002561926841736,
+      "learning_rate": 0.0002,
+      "loss": 1.389,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3137254901960784,
+      "grad_norm": 0.5461082458496094,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3202614379084967,
+      "grad_norm": 0.570399284362793,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2020
+    },
+    {
+      "epoch": 1.326797385620915,
+      "grad_norm": 0.5130975842475891,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.6290071606636047,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3398692810457518,
+      "grad_norm": 0.6165726184844971,
+      "learning_rate": 0.0002,
+      "loss": 1.3873,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34640522875817,
+      "grad_norm": 0.5302083492279053,
+      "learning_rate": 0.0002,
+      "loss": 1.4376,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3529411764705883,
+      "grad_norm": 0.6531406044960022,
+      "learning_rate": 0.0002,
+      "loss": 1.4722,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3594771241830066,
+      "grad_norm": 0.5981236100196838,
+      "learning_rate": 0.0002,
+      "loss": 1.3632,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3660130718954249,
+      "grad_norm": 0.8534150123596191,
+      "learning_rate": 0.0002,
+      "loss": 1.4846,
+      "step": 2090
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 0.695918083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.3249,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3790849673202614,
+      "grad_norm": 0.5830431580543518,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3856209150326797,
+      "grad_norm": 0.5641306638717651,
+      "learning_rate": 0.0002,
+      "loss": 1.5009,
+      "step": 2120
+    },
+    {
+      "epoch": 1.392156862745098,
+      "grad_norm": 0.6354436874389648,
+      "learning_rate": 0.0002,
+      "loss": 1.3985,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3986928104575163,
+      "grad_norm": 0.5707540512084961,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4052287581699345,
+      "grad_norm": 0.7308434844017029,
+      "learning_rate": 0.0002,
+      "loss": 1.3815,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 0.5879750847816467,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4183006535947713,
+      "grad_norm": 0.627909243106842,
+      "learning_rate": 0.0002,
+      "loss": 1.3729,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4248366013071896,
+      "grad_norm": 0.5228193998336792,
+      "learning_rate": 0.0002,
+      "loss": 1.3391,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4313725490196079,
+      "grad_norm": 0.6162880659103394,
+      "learning_rate": 0.0002,
+      "loss": 1.457,
+      "step": 2190
+    },
+    {
+      "epoch": 1.4379084967320261,
+      "grad_norm": 0.751610517501831,
+      "learning_rate": 0.0002,
+      "loss": 1.4052,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 0.5623487234115601,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 0.5293187499046326,
+      "learning_rate": 0.0002,
+      "loss": 1.3795,
+      "step": 2220
+    },
+    {
+      "epoch": 1.457516339869281,
+      "grad_norm": 0.5903629660606384,
+      "learning_rate": 0.0002,
+      "loss": 1.4247,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4640522875816995,
+      "grad_norm": 0.6084659099578857,
+      "learning_rate": 0.0002,
+      "loss": 1.6167,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.5289803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 2250
+    },
+    {
+      "epoch": 1.477124183006536,
+      "grad_norm": 0.49499568343162537,
+      "learning_rate": 0.0002,
+      "loss": 1.3106,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4836601307189543,
+      "grad_norm": 0.7774190306663513,
+      "learning_rate": 0.0002,
+      "loss": 1.3586,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 0.5932538509368896,
+      "learning_rate": 0.0002,
+      "loss": 1.3075,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4967320261437909,
+      "grad_norm": 0.6009492874145508,
+      "learning_rate": 0.0002,
+      "loss": 1.3241,
+      "step": 2290
+    },
+    {
+      "epoch": 1.5032679738562091,
+      "grad_norm": 0.5559343099594116,
+      "learning_rate": 0.0002,
+      "loss": 1.3728,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5098039215686274,
+      "grad_norm": 0.5956196188926697,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5163398692810457,
+      "grad_norm": 0.5624083876609802,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 2320
+    },
+    {
+      "epoch": 1.522875816993464,
+      "grad_norm": 0.7195250391960144,
+      "learning_rate": 0.0002,
+      "loss": 1.4779,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 0.6010490655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.2938,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5359477124183005,
+      "grad_norm": 0.664929211139679,
+      "learning_rate": 0.0002,
+      "loss": 1.4121,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5424836601307188,
+      "grad_norm": 0.5158776640892029,
+      "learning_rate": 0.0002,
+      "loss": 1.4362,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5490196078431373,
+      "grad_norm": 0.5147154927253723,
+      "learning_rate": 0.0002,
+      "loss": 1.2157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 0.6507977843284607,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 2380
+    },
+    {
+      "epoch": 1.5620915032679739,
+      "grad_norm": 0.5193192362785339,
+      "learning_rate": 0.0002,
+      "loss": 1.2786,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 0.5982314944267273,
+      "learning_rate": 0.0002,
+      "loss": 1.3209,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5751633986928104,
+      "grad_norm": 0.49106258153915405,
+      "learning_rate": 0.0002,
+      "loss": 1.3585,
+      "step": 2410
+    },
+    {
+      "epoch": 1.581699346405229,
+      "grad_norm": 0.6459611654281616,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5882352941176472,
+      "grad_norm": 0.7038363218307495,
+      "learning_rate": 0.0002,
+      "loss": 1.3305,
+      "step": 2430
+    },
+    {
+      "epoch": 1.5947712418300655,
+      "grad_norm": 0.5245680212974548,
+      "learning_rate": 0.0002,
+      "loss": 1.3198,
+      "step": 2440
+    },
+    {
+      "epoch": 1.6013071895424837,
+      "grad_norm": 0.6562076210975647,
+      "learning_rate": 0.0002,
+      "loss": 1.4756,
+      "step": 2450
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 0.6491968035697937,
+      "learning_rate": 0.0002,
+      "loss": 1.5635,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6143790849673203,
+      "grad_norm": 0.604034960269928,
+      "learning_rate": 0.0002,
+      "loss": 1.3657,
+      "step": 2470
+    },
+    {
+      "epoch": 1.6209150326797386,
+      "grad_norm": 0.5759671330451965,
+      "learning_rate": 0.0002,
+      "loss": 1.2693,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6274509803921569,
+      "grad_norm": 0.6157698631286621,
+      "learning_rate": 0.0002,
+      "loss": 1.4136,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6339869281045751,
+      "grad_norm": 0.6513794660568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3929,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6405228758169934,
+      "grad_norm": 0.71990966796875,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 0.7316617369651794,
+      "learning_rate": 0.0002,
+      "loss": 1.4356,
+      "step": 2520
+    },
+    {
+      "epoch": 1.65359477124183,
+      "grad_norm": 0.5475177764892578,
+      "learning_rate": 0.0002,
+      "loss": 1.3119,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6601307189542482,
+      "grad_norm": 0.4911293089389801,
+      "learning_rate": 0.0002,
+      "loss": 1.2998,
+      "step": 2540
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.6122882962226868,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 2550
+    },
+    {
+      "epoch": 1.673202614379085,
+      "grad_norm": 0.5735281705856323,
+      "learning_rate": 0.0002,
+      "loss": 1.3099,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6797385620915033,
+      "grad_norm": 0.5046352744102478,
+      "learning_rate": 0.0002,
+      "loss": 1.2205,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 0.6043242812156677,
+      "learning_rate": 0.0002,
+      "loss": 1.3191,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6928104575163399,
+      "grad_norm": 0.5397698283195496,
+      "learning_rate": 0.0002,
+      "loss": 1.3079,
+      "step": 2590
+    },
+    {
+      "epoch": 1.6993464052287581,
+      "grad_norm": 0.8066475987434387,
+      "learning_rate": 0.0002,
+      "loss": 1.4916,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7058823529411766,
+      "grad_norm": 0.52901691198349,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 2610
+    },
+    {
+      "epoch": 1.712418300653595,
+      "grad_norm": 0.7588503956794739,
+      "learning_rate": 0.0002,
+      "loss": 1.409,
+      "step": 2620
+    },
+    {
+      "epoch": 1.7189542483660132,
+      "grad_norm": 0.6012966632843018,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 0.5927302837371826,
+      "learning_rate": 0.0002,
+      "loss": 1.2583,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7320261437908497,
+      "grad_norm": 0.5086990594863892,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 2650
+    },
+    {
+      "epoch": 1.738562091503268,
+      "grad_norm": 0.6000628471374512,
+      "learning_rate": 0.0002,
+      "loss": 1.5452,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7450980392156863,
+      "grad_norm": 0.6560431718826294,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 2670
+    },
+    {
+      "epoch": 1.7516339869281046,
+      "grad_norm": 0.5738165378570557,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7581699346405228,
+      "grad_norm": 0.5576106905937195,
+      "learning_rate": 0.0002,
+      "loss": 1.3766,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.7298802137374878,
+      "learning_rate": 0.0002,
+      "loss": 1.3277,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7712418300653594,
+      "grad_norm": 0.5751826167106628,
+      "learning_rate": 0.0002,
+      "loss": 1.2618,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 0.6069957613945007,
+      "learning_rate": 0.0002,
+      "loss": 1.35,
+      "step": 2720
+    },
+    {
+      "epoch": 1.784313725490196,
+      "grad_norm": 0.7513017654418945,
+      "learning_rate": 0.0002,
+      "loss": 1.3492,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7908496732026142,
+      "grad_norm": 0.6058869957923889,
+      "learning_rate": 0.0002,
+      "loss": 1.2979,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7973856209150327,
+      "grad_norm": 0.6805883049964905,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 2750
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 0.6864324808120728,
+      "learning_rate": 0.0002,
+      "loss": 1.4062,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8104575163398693,
+      "grad_norm": 0.6261002421379089,
+      "learning_rate": 0.0002,
+      "loss": 1.355,
+      "step": 2770
+    },
+    {
+      "epoch": 1.8169934640522876,
+      "grad_norm": 0.532684862613678,
+      "learning_rate": 0.0002,
+      "loss": 1.5145,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8235294117647058,
+      "grad_norm": 0.6209020018577576,
+      "learning_rate": 0.0002,
+      "loss": 1.3248,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8300653594771243,
+      "grad_norm": 0.67111736536026,
+      "learning_rate": 0.0002,
+      "loss": 1.3908,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8366013071895426,
+      "grad_norm": 0.700467586517334,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 2810
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 0.6968029141426086,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 2820
+    },
+    {
+      "epoch": 1.8496732026143792,
+      "grad_norm": 0.6405863761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3943,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8562091503267975,
+      "grad_norm": 0.5192584991455078,
+      "learning_rate": 0.0002,
+      "loss": 1.4035,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8627450980392157,
+      "grad_norm": 0.4888569414615631,
+      "learning_rate": 0.0002,
+      "loss": 1.2745,
+      "step": 2850
+    },
+    {
+      "epoch": 1.869281045751634,
+      "grad_norm": 0.7625455856323242,
+      "learning_rate": 0.0002,
+      "loss": 1.4324,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8758169934640523,
+      "grad_norm": 0.9162808656692505,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 0.5472783446311951,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.5221137404441833,
+      "learning_rate": 0.0002,
+      "loss": 1.3026,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8954248366013071,
+      "grad_norm": 0.49258849024772644,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 2900
+    },
+    {
+      "epoch": 1.9019607843137254,
+      "grad_norm": 0.5260750651359558,
+      "learning_rate": 0.0002,
+      "loss": 1.3503,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9084967320261437,
+      "grad_norm": 0.6583314538002014,
+      "learning_rate": 0.0002,
+      "loss": 1.3381,
+      "step": 2920
+    },
+    {
+      "epoch": 1.915032679738562,
+      "grad_norm": 0.5728915929794312,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 0.7661453485488892,
+      "learning_rate": 0.0002,
+      "loss": 1.3993,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9281045751633987,
+      "grad_norm": 0.7193911075592041,
+      "learning_rate": 0.0002,
+      "loss": 1.428,
+      "step": 2950
+    },
+    {
+      "epoch": 1.934640522875817,
+      "grad_norm": 0.5007768869400024,
+      "learning_rate": 0.0002,
+      "loss": 1.287,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9411764705882353,
+      "grad_norm": 0.626681923866272,
+      "learning_rate": 0.0002,
+      "loss": 1.372,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9477124183006536,
+      "grad_norm": 0.8692840933799744,
+      "learning_rate": 0.0002,
+      "loss": 1.375,
+      "step": 2980
+    },
+    {
+      "epoch": 1.954248366013072,
+      "grad_norm": 0.6388291120529175,
+      "learning_rate": 0.0002,
+      "loss": 1.3292,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 0.7710477113723755,
+      "learning_rate": 0.0002,
+      "loss": 1.4593,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9673202614379086,
+      "grad_norm": 0.641704261302948,
+      "learning_rate": 0.0002,
+      "loss": 1.5228,
+      "step": 3010
+    },
+    {
+      "epoch": 1.973856209150327,
+      "grad_norm": 0.621148943901062,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9803921568627452,
+      "grad_norm": 0.5119547247886658,
+      "learning_rate": 0.0002,
+      "loss": 1.3017,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9869281045751634,
+      "grad_norm": 0.8104137778282166,
+      "learning_rate": 0.0002,
+      "loss": 1.4923,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9934640522875817,
+      "grad_norm": 0.5856240391731262,
+      "learning_rate": 0.0002,
+      "loss": 1.3331,
+      "step": 3050
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5263566374778748,
+      "learning_rate": 0.0002,
+      "loss": 1.4346,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.4276371002197266,
+      "eval_runtime": 30.5759,
+      "eval_samples_per_second": 14.26,
+      "eval_steps_per_second": 1.799,
+      "step": 3060
+    },
+    {
+      "epoch": 2.0065359477124183,
+      "grad_norm": 0.5143898725509644,
+      "learning_rate": 0.0002,
+      "loss": 1.1636,
+      "step": 3070
+    },
+    {
+      "epoch": 2.0130718954248366,
+      "grad_norm": 0.5749367475509644,
+      "learning_rate": 0.0002,
+      "loss": 1.3335,
+      "step": 3080
+    },
+    {
+      "epoch": 2.019607843137255,
+      "grad_norm": 0.5784284472465515,
+      "learning_rate": 0.0002,
+      "loss": 1.2784,
+      "step": 3090
+    },
+    {
+      "epoch": 2.026143790849673,
+      "grad_norm": 0.5933429598808289,
+      "learning_rate": 0.0002,
+      "loss": 1.2463,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0326797385620914,
+      "grad_norm": 0.6748974919319153,
+      "learning_rate": 0.0002,
+      "loss": 1.2984,
+      "step": 3110
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 0.626399576663971,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 3120
+    },
+    {
+      "epoch": 2.045751633986928,
+      "grad_norm": 0.6173238754272461,
+      "learning_rate": 0.0002,
+      "loss": 1.299,
+      "step": 3130
+    },
+    {
+      "epoch": 2.052287581699346,
+      "grad_norm": 0.807790219783783,
+      "learning_rate": 0.0002,
+      "loss": 1.4144,
+      "step": 3140
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.6222215890884399,
+      "learning_rate": 0.0002,
+      "loss": 1.1953,
+      "step": 3150
+    },
+    {
+      "epoch": 2.065359477124183,
+      "grad_norm": 0.5859580636024475,
+      "learning_rate": 0.0002,
+      "loss": 1.4059,
+      "step": 3160
+    },
+    {
+      "epoch": 2.0718954248366015,
+      "grad_norm": 0.581304132938385,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 3170
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 0.9814971089363098,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 3180
+    },
+    {
+      "epoch": 2.084967320261438,
+      "grad_norm": 0.6491848230361938,
+      "learning_rate": 0.0002,
+      "loss": 1.1962,
+      "step": 3190
+    },
+    {
+      "epoch": 2.0915032679738563,
+      "grad_norm": 0.613680362701416,
+      "learning_rate": 0.0002,
+      "loss": 1.3711,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0980392156862746,
+      "grad_norm": 0.7318086624145508,
+      "learning_rate": 0.0002,
+      "loss": 1.2994,
+      "step": 3210
+    },
+    {
+      "epoch": 2.104575163398693,
+      "grad_norm": 0.6025661826133728,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 3220
+    },
+    {
+      "epoch": 2.111111111111111,
+      "grad_norm": 0.6744484305381775,
+      "learning_rate": 0.0002,
+      "loss": 1.1374,
+      "step": 3230
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 0.6062554121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.3273,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1241830065359477,
+      "grad_norm": 0.6801803112030029,
+      "learning_rate": 0.0002,
+      "loss": 1.3404,
+      "step": 3250
+    },
+    {
+      "epoch": 2.130718954248366,
+      "grad_norm": 0.5218925476074219,
+      "learning_rate": 0.0002,
+      "loss": 1.4084,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1372549019607843,
+      "grad_norm": 0.7494263648986816,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1437908496732025,
+      "grad_norm": 0.7858565449714661,
+      "learning_rate": 0.0002,
+      "loss": 1.3059,
+      "step": 3280
+    },
+    {
+      "epoch": 2.150326797385621,
+      "grad_norm": 0.6836692690849304,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3290
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 0.619848370552063,
+      "learning_rate": 0.0002,
+      "loss": 1.1605,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1633986928104574,
+      "grad_norm": 0.5761294364929199,
+      "learning_rate": 0.0002,
+      "loss": 1.3095,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1699346405228757,
+      "grad_norm": 0.4713786542415619,
+      "learning_rate": 0.0002,
+      "loss": 1.2883,
+      "step": 3320
+    },
+    {
+      "epoch": 2.176470588235294,
+      "grad_norm": 0.7613773345947266,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 3330
+    },
+    {
+      "epoch": 2.183006535947712,
+      "grad_norm": 0.6642718315124512,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 3340
+    },
+    {
+      "epoch": 2.189542483660131,
+      "grad_norm": 0.7162188291549683,
+      "learning_rate": 0.0002,
+      "loss": 1.2048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 0.6916783452033997,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 3360
+    },
+    {
+      "epoch": 2.2026143790849675,
+      "grad_norm": 0.7205567955970764,
+      "learning_rate": 0.0002,
+      "loss": 1.3788,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2091503267973858,
+      "grad_norm": 0.6038199067115784,
+      "learning_rate": 0.0002,
+      "loss": 1.2528,
+      "step": 3380
+    },
+    {
+      "epoch": 2.215686274509804,
+      "grad_norm": 0.6284233927726746,
+      "learning_rate": 0.0002,
+      "loss": 1.2079,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.7450672388076782,
+      "learning_rate": 0.0002,
+      "loss": 1.3057,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2287581699346406,
+      "grad_norm": 0.7755052447319031,
+      "learning_rate": 0.0002,
+      "loss": 1.3034,
+      "step": 3410
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 0.9066099524497986,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 3420
+    },
+    {
+      "epoch": 2.241830065359477,
+      "grad_norm": 0.8578207492828369,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 3430
+    },
+    {
+      "epoch": 2.2483660130718954,
+      "grad_norm": 0.5900213718414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3278,
+      "step": 3440
+    },
+    {
+      "epoch": 2.2549019607843137,
+      "grad_norm": 0.7821717262268066,
+      "learning_rate": 0.0002,
+      "loss": 1.3645,
+      "step": 3450
+    },
+    {
+      "epoch": 2.261437908496732,
+      "grad_norm": 0.6263150572776794,
+      "learning_rate": 0.0002,
+      "loss": 1.183,
+      "step": 3460
+    },
+    {
+      "epoch": 2.2679738562091503,
+      "grad_norm": 0.591799259185791,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 0.5999799966812134,
+      "learning_rate": 0.0002,
+      "loss": 1.2198,
+      "step": 3480
+    },
+    {
+      "epoch": 2.281045751633987,
+      "grad_norm": 0.6227319240570068,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 3490
+    },
+    {
+      "epoch": 2.287581699346405,
+      "grad_norm": 0.719412624835968,
+      "learning_rate": 0.0002,
+      "loss": 1.3865,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2941176470588234,
+      "grad_norm": 1.0361769199371338,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 3510
+    },
+    {
+      "epoch": 2.3006535947712417,
+      "grad_norm": 0.5506668090820312,
+      "learning_rate": 0.0002,
+      "loss": 1.4834,
+      "step": 3520
+    },
+    {
+      "epoch": 2.30718954248366,
+      "grad_norm": 0.6886829733848572,
+      "learning_rate": 0.0002,
+      "loss": 1.2273,
+      "step": 3530
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 0.6226346492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.2296,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3202614379084965,
+      "grad_norm": 0.8109908103942871,
+      "learning_rate": 0.0002,
+      "loss": 1.3087,
+      "step": 3550
+    },
+    {
+      "epoch": 2.326797385620915,
+      "grad_norm": 0.8505511283874512,
+      "learning_rate": 0.0002,
+      "loss": 1.3311,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.5763760209083557,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3398692810457518,
+      "grad_norm": 0.6460059881210327,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 3580
+    },
+    {
+      "epoch": 2.34640522875817,
+      "grad_norm": 0.7175343036651611,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 3590
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.6012630462646484,
+      "learning_rate": 0.0002,
+      "loss": 1.2645,
+      "step": 3600
+    },
+    {
+      "epoch": 2.3594771241830066,
+      "grad_norm": 0.6513685584068298,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 3610
+    },
+    {
+      "epoch": 2.366013071895425,
+      "grad_norm": 0.7465183734893799,
+      "learning_rate": 0.0002,
+      "loss": 1.3271,
+      "step": 3620
+    },
+    {
+      "epoch": 2.372549019607843,
+      "grad_norm": 0.6413124203681946,
+      "learning_rate": 0.0002,
+      "loss": 1.3671,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3790849673202614,
+      "grad_norm": 0.7209562063217163,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 3640
+    },
+    {
+      "epoch": 2.3856209150326797,
+      "grad_norm": 0.6427558660507202,
+      "learning_rate": 0.0002,
+      "loss": 1.1616,
+      "step": 3650
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 0.593958854675293,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3986928104575163,
+      "grad_norm": 0.5944608449935913,
+      "learning_rate": 0.0002,
+      "loss": 1.2802,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4052287581699345,
+      "grad_norm": 0.6606248617172241,
+      "learning_rate": 0.0002,
+      "loss": 1.3542,
+      "step": 3680
+    },
+    {
+      "epoch": 2.411764705882353,
+      "grad_norm": 0.5632851719856262,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 3690
+    },
+    {
+      "epoch": 2.418300653594771,
+      "grad_norm": 0.4976513385772705,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4248366013071894,
+      "grad_norm": 0.6318528056144714,
+      "learning_rate": 0.0002,
+      "loss": 1.1404,
+      "step": 3710
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 0.6306707859039307,
+      "learning_rate": 0.0002,
+      "loss": 1.1705,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4379084967320264,
+      "grad_norm": 0.6362553238868713,
+      "learning_rate": 0.0002,
+      "loss": 1.3524,
+      "step": 3730
+    },
+    {
+      "epoch": 2.4444444444444446,
+      "grad_norm": 0.634368896484375,
+      "learning_rate": 0.0002,
+      "loss": 1.2345,
+      "step": 3740
+    },
+    {
+      "epoch": 2.450980392156863,
+      "grad_norm": 0.6623591184616089,
+      "learning_rate": 0.0002,
+      "loss": 1.2515,
+      "step": 3750
+    },
+    {
+      "epoch": 2.457516339869281,
+      "grad_norm": 0.6150440573692322,
+      "learning_rate": 0.0002,
+      "loss": 1.3246,
+      "step": 3760
+    },
+    {
+      "epoch": 2.4640522875816995,
+      "grad_norm": 0.588935911655426,
+      "learning_rate": 0.0002,
+      "loss": 1.2666,
+      "step": 3770
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 0.7388206124305725,
+      "learning_rate": 0.0002,
+      "loss": 1.3918,
+      "step": 3780
+    },
+    {
+      "epoch": 2.477124183006536,
+      "grad_norm": 0.621825098991394,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4836601307189543,
+      "grad_norm": 0.7691677212715149,
+      "learning_rate": 0.0002,
+      "loss": 1.359,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4901960784313726,
+      "grad_norm": 1.1661969423294067,
+      "learning_rate": 0.0002,
+      "loss": 1.3399,
+      "step": 3810
+    },
+    {
+      "epoch": 2.496732026143791,
+      "grad_norm": 0.6837884187698364,
+      "learning_rate": 0.0002,
+      "loss": 1.461,
+      "step": 3820
+    },
+    {
+      "epoch": 2.503267973856209,
+      "grad_norm": 0.6978904008865356,
+      "learning_rate": 0.0002,
+      "loss": 1.2823,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 0.6121411323547363,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5163398692810457,
+      "grad_norm": 0.7813326120376587,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 3850
+    },
+    {
+      "epoch": 2.522875816993464,
+      "grad_norm": 0.5390260219573975,
+      "learning_rate": 0.0002,
+      "loss": 1.1543,
+      "step": 3860
+    },
+    {
+      "epoch": 2.5294117647058822,
+      "grad_norm": 0.8283252716064453,
+      "learning_rate": 0.0002,
+      "loss": 1.2032,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5359477124183005,
+      "grad_norm": 0.8527186512947083,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 3880
+    },
+    {
+      "epoch": 2.542483660130719,
+      "grad_norm": 0.8405382633209229,
+      "learning_rate": 0.0002,
+      "loss": 1.3469,
+      "step": 3890
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 0.5650738477706909,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 3900
+    },
+    {
+      "epoch": 2.5555555555555554,
+      "grad_norm": 0.620121955871582,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 3910
+    },
+    {
+      "epoch": 2.5620915032679736,
+      "grad_norm": 0.5983527898788452,
+      "learning_rate": 0.0002,
+      "loss": 1.2524,
+      "step": 3920
+    },
+    {
+      "epoch": 2.568627450980392,
+      "grad_norm": 0.686623215675354,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 3930
+    },
+    {
+      "epoch": 2.57516339869281,
+      "grad_norm": 0.6805831789970398,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 3940
+    },
+    {
+      "epoch": 2.581699346405229,
+      "grad_norm": 0.6994825601577759,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 3950
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 0.728549599647522,
+      "learning_rate": 0.0002,
+      "loss": 1.3446,
+      "step": 3960
+    },
+    {
+      "epoch": 2.5947712418300655,
+      "grad_norm": 0.775236964225769,
+      "learning_rate": 0.0002,
+      "loss": 1.4039,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6013071895424837,
+      "grad_norm": 0.5057447552680969,
+      "learning_rate": 0.0002,
+      "loss": 1.2742,
+      "step": 3980
+    },
+    {
+      "epoch": 2.607843137254902,
+      "grad_norm": 0.6564450263977051,
+      "learning_rate": 0.0002,
+      "loss": 1.2764,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6143790849673203,
+      "grad_norm": 0.5342249870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.3269,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6209150326797386,
+      "grad_norm": 0.5508961081504822,
+      "learning_rate": 0.0002,
+      "loss": 1.3102,
+      "step": 4010
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 0.5716235637664795,
+      "learning_rate": 0.0002,
+      "loss": 1.3636,
+      "step": 4020
+    },
+    {
+      "epoch": 2.633986928104575,
+      "grad_norm": 0.8049232363700867,
+      "learning_rate": 0.0002,
+      "loss": 1.3465,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6405228758169934,
+      "grad_norm": 0.5574354529380798,
+      "learning_rate": 0.0002,
+      "loss": 1.2342,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.6302093863487244,
+      "learning_rate": 0.0002,
+      "loss": 1.2419,
+      "step": 4050
+    },
+    {
+      "epoch": 2.65359477124183,
+      "grad_norm": 1.1868736743927002,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 4060
+    },
+    {
+      "epoch": 2.6601307189542482,
+      "grad_norm": 0.6738120317459106,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 4070
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.6614423990249634,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 4080
+    },
+    {
+      "epoch": 2.6732026143790852,
+      "grad_norm": 0.7297604084014893,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6797385620915035,
+      "grad_norm": 0.9421682357788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1596,
+      "step": 4100
+    },
+    {
+      "epoch": 2.686274509803922,
+      "grad_norm": 0.5286222696304321,
+      "learning_rate": 0.0002,
+      "loss": 1.3002,
+      "step": 4110
+    },
+    {
+      "epoch": 2.69281045751634,
+      "grad_norm": 0.6849271655082703,
+      "learning_rate": 0.0002,
+      "loss": 1.3936,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6993464052287583,
+      "grad_norm": 0.6811320185661316,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 0.4968419373035431,
+      "learning_rate": 0.0002,
+      "loss": 1.2897,
+      "step": 4140
+    },
+    {
+      "epoch": 2.712418300653595,
+      "grad_norm": 0.8074267506599426,
+      "learning_rate": 0.0002,
+      "loss": 1.3322,
+      "step": 4150
+    },
+    {
+      "epoch": 2.718954248366013,
+      "grad_norm": 0.6756376028060913,
+      "learning_rate": 0.0002,
+      "loss": 1.1759,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7254901960784315,
+      "grad_norm": 0.6921583414077759,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4170
+    },
+    {
+      "epoch": 2.7320261437908497,
+      "grad_norm": 0.7049834132194519,
+      "learning_rate": 0.0002,
+      "loss": 1.3413,
+      "step": 4180
+    },
+    {
+      "epoch": 2.738562091503268,
+      "grad_norm": 0.7011390328407288,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 0.6977843642234802,
+      "learning_rate": 0.0002,
+      "loss": 1.2364,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7516339869281046,
+      "grad_norm": 0.6717000603675842,
+      "learning_rate": 0.0002,
+      "loss": 1.2533,
+      "step": 4210
+    },
+    {
+      "epoch": 2.758169934640523,
+      "grad_norm": 1.0223724842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 4220
+    },
+    {
+      "epoch": 2.764705882352941,
+      "grad_norm": 0.6573330760002136,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4230
+    },
+    {
+      "epoch": 2.7712418300653594,
+      "grad_norm": 0.6684938073158264,
+      "learning_rate": 0.0002,
+      "loss": 1.4219,
+      "step": 4240
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.7426793575286865,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 4250
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 0.557826578617096,
+      "learning_rate": 0.0002,
+      "loss": 1.2904,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7908496732026142,
+      "grad_norm": 0.6669870018959045,
+      "learning_rate": 0.0002,
+      "loss": 1.3262,
+      "step": 4270
+    },
+    {
+      "epoch": 2.7973856209150325,
+      "grad_norm": 0.5349969267845154,
+      "learning_rate": 0.0002,
+      "loss": 1.2369,
+      "step": 4280
+    },
+    {
+      "epoch": 2.803921568627451,
+      "grad_norm": 0.7262802124023438,
+      "learning_rate": 0.0002,
+      "loss": 1.3769,
+      "step": 4290
+    },
+    {
+      "epoch": 2.810457516339869,
+      "grad_norm": 0.768211841583252,
+      "learning_rate": 0.0002,
+      "loss": 1.3373,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8169934640522873,
+      "grad_norm": 0.5958252549171448,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 0.8451310396194458,
+      "learning_rate": 0.0002,
+      "loss": 1.4113,
+      "step": 4320
+    },
+    {
+      "epoch": 2.8300653594771243,
+      "grad_norm": 0.6544435024261475,
+      "learning_rate": 0.0002,
+      "loss": 1.2454,
+      "step": 4330
+    },
+    {
+      "epoch": 2.8366013071895426,
+      "grad_norm": 0.6177433133125305,
+      "learning_rate": 0.0002,
+      "loss": 1.2777,
+      "step": 4340
+    },
+    {
+      "epoch": 2.843137254901961,
+      "grad_norm": 0.6324988007545471,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 4350
+    },
+    {
+      "epoch": 2.849673202614379,
+      "grad_norm": 0.6884300708770752,
+      "learning_rate": 0.0002,
+      "loss": 1.4117,
+      "step": 4360
+    },
+    {
+      "epoch": 2.8562091503267975,
+      "grad_norm": 0.8952897191047668,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 4370
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 1.0260103940963745,
+      "learning_rate": 0.0002,
+      "loss": 1.2814,
+      "step": 4380
+    },
+    {
+      "epoch": 2.869281045751634,
+      "grad_norm": 0.9134647250175476,
+      "learning_rate": 0.0002,
+      "loss": 1.2893,
+      "step": 4390
+    },
+    {
+      "epoch": 2.8758169934640523,
+      "grad_norm": 0.5637717843055725,
+      "learning_rate": 0.0002,
+      "loss": 1.171,
+      "step": 4400
+    },
+    {
+      "epoch": 2.8823529411764706,
+      "grad_norm": 0.7530393004417419,
+      "learning_rate": 0.0002,
+      "loss": 1.3422,
+      "step": 4410
+    },
+    {
+      "epoch": 2.888888888888889,
+      "grad_norm": 0.7202680706977844,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 4420
+    },
+    {
+      "epoch": 2.895424836601307,
+      "grad_norm": 0.7177144885063171,
+      "learning_rate": 0.0002,
+      "loss": 1.2913,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 0.5996816754341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 4440
+    },
+    {
+      "epoch": 2.9084967320261437,
+      "grad_norm": 0.6542447209358215,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 4450
+    },
+    {
+      "epoch": 2.915032679738562,
+      "grad_norm": 1.0753740072250366,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9215686274509802,
+      "grad_norm": 0.6956136226654053,
+      "learning_rate": 0.0002,
+      "loss": 1.3193,
+      "step": 4470
+    },
+    {
+      "epoch": 2.928104575163399,
+      "grad_norm": 0.7702530026435852,
+      "learning_rate": 0.0002,
+      "loss": 1.2486,
+      "step": 4480
+    },
+    {
+      "epoch": 2.9346405228758172,
+      "grad_norm": 0.7763232588768005,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 4490
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.6393085718154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1647,
+      "step": 4500
+    },
+    {
+      "epoch": 2.947712418300654,
+      "grad_norm": 0.987770676612854,
+      "learning_rate": 0.0002,
+      "loss": 1.211,
+      "step": 4510
+    },
+    {
+      "epoch": 2.954248366013072,
+      "grad_norm": 0.5995016098022461,
+      "learning_rate": 0.0002,
+      "loss": 1.1529,
+      "step": 4520
+    },
+    {
+      "epoch": 2.9607843137254903,
+      "grad_norm": 0.745650053024292,
+      "learning_rate": 0.0002,
+      "loss": 1.2358,
+      "step": 4530
+    },
+    {
+      "epoch": 2.9673202614379086,
+      "grad_norm": 0.7429282069206238,
+      "learning_rate": 0.0002,
+      "loss": 1.2115,
+      "step": 4540
+    },
+    {
+      "epoch": 2.973856209150327,
+      "grad_norm": 0.5927486419677734,
+      "learning_rate": 0.0002,
+      "loss": 1.2262,
+      "step": 4550
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 0.6775153875350952,
+      "learning_rate": 0.0002,
+      "loss": 1.3173,
+      "step": 4560
+    },
+    {
+      "epoch": 2.9869281045751634,
+      "grad_norm": 0.7128435373306274,
+      "learning_rate": 0.0002,
+      "loss": 1.279,
+      "step": 4570
+    },
+    {
+      "epoch": 2.9934640522875817,
+      "grad_norm": 0.7470937967300415,
+      "learning_rate": 0.0002,
+      "loss": 1.2451,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9295375943183899,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.4131312370300293,
+      "eval_runtime": 31.8967,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 1.724,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0065359477124183,
+      "grad_norm": 0.6926420331001282,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0130718954248366,
+      "grad_norm": 0.6656355857849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 4610
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 0.9901936650276184,
+      "learning_rate": 0.0002,
+      "loss": 1.308,
+      "step": 4620
+    },
+    {
+      "epoch": 3.026143790849673,
+      "grad_norm": 0.6713474988937378,
+      "learning_rate": 0.0002,
+      "loss": 1.22,
+      "step": 4630
+    },
+    {
+      "epoch": 3.0326797385620914,
+      "grad_norm": 0.6199324131011963,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 4640
+    },
+    {
+      "epoch": 3.0392156862745097,
+      "grad_norm": 0.7180785536766052,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 4650
+    },
+    {
+      "epoch": 3.045751633986928,
+      "grad_norm": 0.8256588578224182,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 4660
+    },
+    {
+      "epoch": 3.052287581699346,
+      "grad_norm": 0.6637389063835144,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 4670
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 0.6980698108673096,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 4680
+    },
+    {
+      "epoch": 3.065359477124183,
+      "grad_norm": 0.8091534972190857,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 4690
+    },
+    {
+      "epoch": 3.0718954248366015,
+      "grad_norm": 0.5715174078941345,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 4700
+    },
+    {
+      "epoch": 3.0784313725490198,
+      "grad_norm": 0.735639750957489,
+      "learning_rate": 0.0002,
+      "loss": 1.1427,
+      "step": 4710
+    },
+    {
+      "epoch": 3.084967320261438,
+      "grad_norm": 0.7619708180427551,
+      "learning_rate": 0.0002,
+      "loss": 1.1522,
+      "step": 4720
+    },
+    {
+      "epoch": 3.0915032679738563,
+      "grad_norm": 1.263566017150879,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 4730
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 0.6600871682167053,
+      "learning_rate": 0.0002,
+      "loss": 1.1348,
+      "step": 4740
+    },
+    {
+      "epoch": 3.104575163398693,
+      "grad_norm": 0.717792809009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1766,
+      "step": 4750
+    },
+    {
+      "epoch": 3.111111111111111,
+      "grad_norm": 0.853714644908905,
+      "learning_rate": 0.0002,
+      "loss": 1.088,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1176470588235294,
+      "grad_norm": 1.1004153490066528,
+      "learning_rate": 0.0002,
+      "loss": 1.2031,
+      "step": 4770
+    },
+    {
+      "epoch": 3.1241830065359477,
+      "grad_norm": 0.8566235899925232,
+      "learning_rate": 0.0002,
+      "loss": 1.3295,
+      "step": 4780
+    },
+    {
+      "epoch": 3.130718954248366,
+      "grad_norm": 0.8315296173095703,
+      "learning_rate": 0.0002,
+      "loss": 1.2436,
+      "step": 4790
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 0.8020524978637695,
+      "learning_rate": 0.0002,
+      "loss": 1.32,
+      "step": 4800
+    },
+    {
+      "epoch": 3.1437908496732025,
+      "grad_norm": 0.7564275860786438,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 4810
+    },
+    {
+      "epoch": 3.150326797385621,
+      "grad_norm": 0.9077776670455933,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 4820
+    },
+    {
+      "epoch": 3.156862745098039,
+      "grad_norm": 0.6323099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.1399,
+      "step": 4830
+    },
+    {
+      "epoch": 3.1633986928104574,
+      "grad_norm": 0.6625368595123291,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 4840
+    },
+    {
+      "epoch": 3.1699346405228757,
+      "grad_norm": 0.8119261860847473,
+      "learning_rate": 0.0002,
+      "loss": 1.066,
+      "step": 4850
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 0.6399450898170471,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 4860
+    },
+    {
+      "epoch": 3.183006535947712,
+      "grad_norm": 1.0659016370773315,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 4870
+    },
+    {
+      "epoch": 3.189542483660131,
+      "grad_norm": 0.8040369749069214,
+      "learning_rate": 0.0002,
+      "loss": 1.2914,
+      "step": 4880
+    },
+    {
+      "epoch": 3.196078431372549,
+      "grad_norm": 0.7784733176231384,
+      "learning_rate": 0.0002,
+      "loss": 1.1996,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2026143790849675,
+      "grad_norm": 0.9660294651985168,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 4900
+    },
+    {
+      "epoch": 3.2091503267973858,
+      "grad_norm": 1.0676977634429932,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 4910
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 0.5877565741539001,
+      "learning_rate": 0.0002,
+      "loss": 1.0083,
+      "step": 4920
+    },
+    {
+      "epoch": 3.2222222222222223,
+      "grad_norm": 0.6164032816886902,
+      "learning_rate": 0.0002,
+      "loss": 1.1046,
+      "step": 4930
+    },
+    {
+      "epoch": 3.2287581699346406,
+      "grad_norm": 0.7627606987953186,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 4940
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.7442803978919983,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 4950
+    },
+    {
+      "epoch": 3.241830065359477,
+      "grad_norm": 0.7277812361717224,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 4960
+    },
+    {
+      "epoch": 3.2483660130718954,
+      "grad_norm": 1.0301902294158936,
+      "learning_rate": 0.0002,
+      "loss": 1.2237,
+      "step": 4970
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 0.7798232436180115,
+      "learning_rate": 0.0002,
+      "loss": 1.1466,
+      "step": 4980
+    },
+    {
+      "epoch": 3.261437908496732,
+      "grad_norm": 1.210265874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.2142,
+      "step": 4990
+    },
+    {
+      "epoch": 3.2679738562091503,
+      "grad_norm": 0.6677713990211487,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2745098039215685,
+      "grad_norm": 1.0524500608444214,
+      "learning_rate": 0.0002,
+      "loss": 1.3294,
+      "step": 5010
+    },
+    {
+      "epoch": 3.281045751633987,
+      "grad_norm": 0.7091745734214783,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 5020
+    },
+    {
+      "epoch": 3.287581699346405,
+      "grad_norm": 0.8523224592208862,
+      "learning_rate": 0.0002,
+      "loss": 1.1891,
+      "step": 5030
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 0.6120608448982239,
+      "learning_rate": 0.0002,
+      "loss": 1.1925,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3006535947712417,
+      "grad_norm": 0.7437472939491272,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 5050
+    },
+    {
+      "epoch": 3.30718954248366,
+      "grad_norm": 0.7611715197563171,
+      "learning_rate": 0.0002,
+      "loss": 1.1295,
+      "step": 5060
+    },
+    {
+      "epoch": 3.313725490196078,
+      "grad_norm": 0.7249704003334045,
+      "learning_rate": 0.0002,
+      "loss": 1.0531,
+      "step": 5070
+    },
+    {
+      "epoch": 3.3202614379084965,
+      "grad_norm": 0.7316247820854187,
+      "learning_rate": 0.0002,
+      "loss": 1.2292,
+      "step": 5080
+    },
+    {
+      "epoch": 3.326797385620915,
+      "grad_norm": 0.562412440776825,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 5090
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.7052176594734192,
+      "learning_rate": 0.0002,
+      "loss": 1.0736,
+      "step": 5100
+    },
+    {
+      "epoch": 3.3398692810457518,
+      "grad_norm": 0.7714211344718933,
+      "learning_rate": 0.0002,
+      "loss": 1.122,
+      "step": 5110
+    },
+    {
+      "epoch": 3.34640522875817,
+      "grad_norm": 1.0436055660247803,
+      "learning_rate": 0.0002,
+      "loss": 1.1684,
+      "step": 5120
+    },
+    {
+      "epoch": 3.3529411764705883,
+      "grad_norm": 0.8867271542549133,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 5130
+    },
+    {
+      "epoch": 3.3594771241830066,
+      "grad_norm": 0.8371267914772034,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 5140
+    },
+    {
+      "epoch": 3.366013071895425,
+      "grad_norm": 0.7257837057113647,
+      "learning_rate": 0.0002,
+      "loss": 1.1073,
+      "step": 5150
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 0.7102002501487732,
+      "learning_rate": 0.0002,
+      "loss": 1.1162,
+      "step": 5160
+    },
+    {
+      "epoch": 3.3790849673202614,
+      "grad_norm": 0.7636350393295288,
+      "learning_rate": 0.0002,
+      "loss": 1.2056,
+      "step": 5170
+    },
+    {
+      "epoch": 3.3856209150326797,
+      "grad_norm": 0.6887359619140625,
+      "learning_rate": 0.0002,
+      "loss": 1.0708,
+      "step": 5180
+    },
+    {
+      "epoch": 3.392156862745098,
+      "grad_norm": 0.8141424655914307,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 5190
+    },
+    {
+      "epoch": 3.3986928104575163,
+      "grad_norm": 0.694423496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5200
+    },
+    {
+      "epoch": 3.4052287581699345,
+      "grad_norm": 0.914013683795929,
+      "learning_rate": 0.0002,
+      "loss": 1.2945,
+      "step": 5210
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 0.8503239750862122,
+      "learning_rate": 0.0002,
+      "loss": 1.1413,
+      "step": 5220
+    },
+    {
+      "epoch": 3.418300653594771,
+      "grad_norm": 0.6196836233139038,
+      "learning_rate": 0.0002,
+      "loss": 1.2696,
+      "step": 5230
+    },
+    {
+      "epoch": 3.4248366013071894,
+      "grad_norm": 1.0760811567306519,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 5240
+    },
+    {
+      "epoch": 3.431372549019608,
+      "grad_norm": 0.6524698138237,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 5250
+    },
+    {
+      "epoch": 3.4379084967320264,
+      "grad_norm": 0.674467921257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 5260
+    },
+    {
+      "epoch": 3.4444444444444446,
+      "grad_norm": 0.7690372467041016,
+      "learning_rate": 0.0002,
+      "loss": 1.1015,
+      "step": 5270
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 0.8751813769340515,
+      "learning_rate": 0.0002,
+      "loss": 1.2511,
+      "step": 5280
+    },
+    {
+      "epoch": 3.457516339869281,
+      "grad_norm": 0.750407874584198,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 5290
+    },
+    {
+      "epoch": 3.4640522875816995,
+      "grad_norm": 0.5991823077201843,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 5300
+    },
+    {
+      "epoch": 3.4705882352941178,
+      "grad_norm": 1.0164772272109985,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 5310
+    },
+    {
+      "epoch": 3.477124183006536,
+      "grad_norm": 0.8704105019569397,
+      "learning_rate": 0.0002,
+      "loss": 1.2354,
+      "step": 5320
+    },
+    {
+      "epoch": 3.4836601307189543,
+      "grad_norm": 0.709102213382721,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 5330
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 0.6273632049560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2425,
+      "step": 5340
+    },
+    {
+      "epoch": 3.496732026143791,
+      "grad_norm": 0.6807359457015991,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 5350
+    },
+    {
+      "epoch": 3.503267973856209,
+      "grad_norm": 0.7085188627243042,
+      "learning_rate": 0.0002,
+      "loss": 1.131,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5098039215686274,
+      "grad_norm": 0.6938307881355286,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 5370
+    },
+    {
+      "epoch": 3.5163398692810457,
+      "grad_norm": 0.8544146418571472,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 5380
+    },
+    {
+      "epoch": 3.522875816993464,
+      "grad_norm": 0.7889642119407654,
+      "learning_rate": 0.0002,
+      "loss": 1.2181,
+      "step": 5390
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.7858421206474304,
+      "learning_rate": 0.0002,
+      "loss": 1.1691,
+      "step": 5400
+    },
+    {
+      "epoch": 3.5359477124183005,
+      "grad_norm": 0.8547123074531555,
+      "learning_rate": 0.0002,
+      "loss": 1.2374,
+      "step": 5410
+    },
+    {
+      "epoch": 3.542483660130719,
+      "grad_norm": 0.8218181133270264,
+      "learning_rate": 0.0002,
+      "loss": 1.196,
+      "step": 5420
+    },
+    {
+      "epoch": 3.549019607843137,
+      "grad_norm": 1.153623342514038,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 5430
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 1.1321099996566772,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 5440
+    },
+    {
+      "epoch": 3.5620915032679736,
+      "grad_norm": 0.9495334029197693,
+      "learning_rate": 0.0002,
+      "loss": 1.2224,
+      "step": 5450
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 0.8743821978569031,
+      "learning_rate": 0.0002,
+      "loss": 1.2869,
+      "step": 5460
+    },
+    {
+      "epoch": 3.57516339869281,
+      "grad_norm": 0.7513086795806885,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 5470
+    },
+    {
+      "epoch": 3.581699346405229,
+      "grad_norm": 1.0139480829238892,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 5480
+    },
+    {
+      "epoch": 3.588235294117647,
+      "grad_norm": 0.6615135073661804,
+      "learning_rate": 0.0002,
+      "loss": 1.1706,
+      "step": 5490
+    },
+    {
+      "epoch": 3.5947712418300655,
+      "grad_norm": 1.180798888206482,
+      "learning_rate": 0.0002,
+      "loss": 1.3906,
+      "step": 5500
+    },
+    {
+      "epoch": 3.6013071895424837,
+      "grad_norm": 0.7085279226303101,
+      "learning_rate": 0.0002,
+      "loss": 1.2391,
+      "step": 5510
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 0.540268063545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1623,
+      "step": 5520
+    },
+    {
+      "epoch": 3.6143790849673203,
+      "grad_norm": 0.7905671000480652,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 5530
+    },
+    {
+      "epoch": 3.6209150326797386,
+      "grad_norm": 0.8457717299461365,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 5540
+    },
+    {
+      "epoch": 3.627450980392157,
+      "grad_norm": 0.7102677822113037,
+      "learning_rate": 0.0002,
+      "loss": 1.1799,
+      "step": 5550
+    },
+    {
+      "epoch": 3.633986928104575,
+      "grad_norm": 0.7179514765739441,
+      "learning_rate": 0.0002,
+      "loss": 1.2394,
+      "step": 5560
+    },
+    {
+      "epoch": 3.6405228758169934,
+      "grad_norm": 1.0854148864746094,
+      "learning_rate": 0.0002,
+      "loss": 1.2019,
+      "step": 5570
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 0.8209951519966125,
+      "learning_rate": 0.0002,
+      "loss": 1.1986,
+      "step": 5580
+    },
+    {
+      "epoch": 3.65359477124183,
+      "grad_norm": 0.6944138407707214,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 5590
+    },
+    {
+      "epoch": 3.6601307189542482,
+      "grad_norm": 0.7675473093986511,
+      "learning_rate": 0.0002,
+      "loss": 1.3226,
+      "step": 5600
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6683364510536194,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 5610
+    },
+    {
+      "epoch": 3.6732026143790852,
+      "grad_norm": 0.7920727133750916,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 5620
+    },
+    {
+      "epoch": 3.6797385620915035,
+      "grad_norm": 0.9440218806266785,
+      "learning_rate": 0.0002,
+      "loss": 1.2287,
+      "step": 5630
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 0.6600824594497681,
+      "learning_rate": 0.0002,
+      "loss": 1.2444,
+      "step": 5640
+    },
+    {
+      "epoch": 3.69281045751634,
+      "grad_norm": 0.6860619187355042,
+      "learning_rate": 0.0002,
+      "loss": 1.191,
+      "step": 5650
+    },
+    {
+      "epoch": 3.6993464052287583,
+      "grad_norm": 0.6579713225364685,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 5660
+    },
+    {
+      "epoch": 3.7058823529411766,
+      "grad_norm": 0.661081075668335,
+      "learning_rate": 0.0002,
+      "loss": 1.1464,
+      "step": 5670
+    },
+    {
+      "epoch": 3.712418300653595,
+      "grad_norm": 1.0968825817108154,
+      "learning_rate": 0.0002,
+      "loss": 1.289,
+      "step": 5680
+    },
+    {
+      "epoch": 3.718954248366013,
+      "grad_norm": 0.8066844940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.192,
+      "step": 5690
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 0.8341682553291321,
+      "learning_rate": 0.0002,
+      "loss": 1.2322,
+      "step": 5700
+    },
+    {
+      "epoch": 3.7320261437908497,
+      "grad_norm": 0.6682852506637573,
+      "learning_rate": 0.0002,
+      "loss": 1.1473,
+      "step": 5710
+    },
+    {
+      "epoch": 3.738562091503268,
+      "grad_norm": 0.898595929145813,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 5720
+    },
+    {
+      "epoch": 3.7450980392156863,
+      "grad_norm": 0.6876054406166077,
+      "learning_rate": 0.0002,
+      "loss": 1.0919,
+      "step": 5730
+    },
+    {
+      "epoch": 3.7516339869281046,
+      "grad_norm": 0.7817103266716003,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 5740
+    },
+    {
+      "epoch": 3.758169934640523,
+      "grad_norm": 0.5840168595314026,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 5750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 0.6263918876647949,
+      "learning_rate": 0.0002,
+      "loss": 1.1279,
+      "step": 5760
+    },
+    {
+      "epoch": 3.7712418300653594,
+      "grad_norm": 0.7948952317237854,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 5770
+    },
+    {
+      "epoch": 3.7777777777777777,
+      "grad_norm": 0.6700998544692993,
+      "learning_rate": 0.0002,
+      "loss": 1.149,
+      "step": 5780
+    },
+    {
+      "epoch": 3.784313725490196,
+      "grad_norm": 1.1169519424438477,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 5790
+    },
+    {
+      "epoch": 3.7908496732026142,
+      "grad_norm": 0.8354471325874329,
+      "learning_rate": 0.0002,
+      "loss": 1.064,
+      "step": 5800
+    },
+    {
+      "epoch": 3.7973856209150325,
+      "grad_norm": 0.6304181814193726,
+      "learning_rate": 0.0002,
+      "loss": 1.2104,
+      "step": 5810
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 0.6919655799865723,
+      "learning_rate": 0.0002,
+      "loss": 1.2059,
+      "step": 5820
+    },
+    {
+      "epoch": 3.810457516339869,
+      "grad_norm": 0.600385844707489,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5830
+    },
+    {
+      "epoch": 3.8169934640522873,
+      "grad_norm": 0.8406319618225098,
+      "learning_rate": 0.0002,
+      "loss": 1.2324,
+      "step": 5840
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.7594282031059265,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5850
+    },
+    {
+      "epoch": 3.8300653594771243,
+      "grad_norm": 0.8179879784584045,
+      "learning_rate": 0.0002,
+      "loss": 1.1903,
+      "step": 5860
+    },
+    {
+      "epoch": 3.8366013071895426,
+      "grad_norm": 1.141430377960205,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 5870
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 0.6595550775527954,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 5880
+    },
+    {
+      "epoch": 3.849673202614379,
+      "grad_norm": 0.7499435544013977,
+      "learning_rate": 0.0002,
+      "loss": 1.2378,
+      "step": 5890
+    },
+    {
+      "epoch": 3.8562091503267975,
+      "grad_norm": 0.7851517200469971,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 5900
+    },
+    {
+      "epoch": 3.8627450980392157,
+      "grad_norm": 1.0533545017242432,
+      "learning_rate": 0.0002,
+      "loss": 1.162,
+      "step": 5910
+    },
+    {
+      "epoch": 3.869281045751634,
+      "grad_norm": 0.960086464881897,
+      "learning_rate": 0.0002,
+      "loss": 1.3576,
+      "step": 5920
+    },
+    {
+      "epoch": 3.8758169934640523,
+      "grad_norm": 0.9952049851417542,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 5930
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 0.7884191274642944,
+      "learning_rate": 0.0002,
+      "loss": 1.2027,
+      "step": 5940
+    },
+    {
+      "epoch": 3.888888888888889,
+      "grad_norm": 0.7461766600608826,
+      "learning_rate": 0.0002,
+      "loss": 1.1796,
+      "step": 5950
+    },
+    {
+      "epoch": 3.895424836601307,
+      "grad_norm": 0.9594355821609497,
+      "learning_rate": 0.0002,
+      "loss": 1.2251,
+      "step": 5960
+    },
+    {
+      "epoch": 3.9019607843137254,
+      "grad_norm": 0.8179471492767334,
+      "learning_rate": 0.0002,
+      "loss": 1.1164,
+      "step": 5970
+    },
+    {
+      "epoch": 3.9084967320261437,
+      "grad_norm": 0.8240267634391785,
+      "learning_rate": 0.0002,
+      "loss": 1.2421,
+      "step": 5980
+    },
+    {
+      "epoch": 3.915032679738562,
+      "grad_norm": 0.7462618350982666,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 5990
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 0.711207389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.2124,
+      "step": 6000
+    },
+    {
+      "epoch": 3.928104575163399,
+      "grad_norm": 0.6910956501960754,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 6010
+    },
+    {
+      "epoch": 3.9346405228758172,
+      "grad_norm": 0.749093770980835,
+      "learning_rate": 0.0002,
+      "loss": 1.2127,
+      "step": 6020
+    },
+    {
+      "epoch": 3.9411764705882355,
+      "grad_norm": 1.3332762718200684,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 6030
+    },
+    {
+      "epoch": 3.947712418300654,
+      "grad_norm": 0.71457439661026,
+      "learning_rate": 0.0002,
+      "loss": 1.1442,
+      "step": 6040
+    },
+    {
+      "epoch": 3.954248366013072,
+      "grad_norm": 1.1205238103866577,
+      "learning_rate": 0.0002,
+      "loss": 1.339,
+      "step": 6050
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 0.6958928108215332,
+      "learning_rate": 0.0002,
+      "loss": 1.2962,
+      "step": 6060
+    },
+    {
+      "epoch": 3.9673202614379086,
+      "grad_norm": 0.7518056035041809,
+      "learning_rate": 0.0002,
+      "loss": 1.1802,
+      "step": 6070
+    },
+    {
+      "epoch": 3.973856209150327,
+      "grad_norm": 0.8010755777359009,
+      "learning_rate": 0.0002,
+      "loss": 1.1179,
+      "step": 6080
+    },
+    {
+      "epoch": 3.980392156862745,
+      "grad_norm": 0.7492658495903015,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 6090
+    },
+    {
+      "epoch": 3.9869281045751634,
+      "grad_norm": 0.900704562664032,
+      "learning_rate": 0.0002,
+      "loss": 1.2113,
+      "step": 6100
+    },
+    {
+      "epoch": 3.9934640522875817,
+      "grad_norm": 0.7997331619262695,
+      "learning_rate": 0.0002,
+      "loss": 1.1106,
+      "step": 6110
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7163209319114685,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 6120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.4113320112228394,
+      "eval_runtime": 33.7199,
+      "eval_samples_per_second": 12.93,
+      "eval_steps_per_second": 1.631,
+      "step": 6120
+    },
+    {
+      "epoch": 4.006535947712418,
+      "grad_norm": 0.9527022838592529,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 6130
+    },
+    {
+      "epoch": 4.0130718954248366,
+      "grad_norm": 0.7603210210800171,
+      "learning_rate": 0.0002,
+      "loss": 1.101,
+      "step": 6140
+    },
+    {
+      "epoch": 4.019607843137255,
+      "grad_norm": 1.127387523651123,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 6150
+    },
+    {
+      "epoch": 4.026143790849673,
+      "grad_norm": 0.8290133476257324,
+      "learning_rate": 0.0002,
+      "loss": 1.0734,
+      "step": 6160
+    },
+    {
+      "epoch": 4.032679738562091,
+      "grad_norm": 0.9912241101264954,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 6170
+    },
+    {
+      "epoch": 4.03921568627451,
+      "grad_norm": 0.947005033493042,
+      "learning_rate": 0.0002,
+      "loss": 1.0719,
+      "step": 6180
+    },
+    {
+      "epoch": 4.045751633986928,
+      "grad_norm": 0.707466185092926,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 6190
+    },
+    {
+      "epoch": 4.052287581699346,
+      "grad_norm": 1.0604327917099,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6200
+    },
+    {
+      "epoch": 4.0588235294117645,
+      "grad_norm": 0.7848685383796692,
+      "learning_rate": 0.0002,
+      "loss": 1.0375,
+      "step": 6210
+    },
+    {
+      "epoch": 4.065359477124183,
+      "grad_norm": 0.8475256562232971,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 6220
+    },
+    {
+      "epoch": 4.071895424836601,
+      "grad_norm": 0.9759448766708374,
+      "learning_rate": 0.0002,
+      "loss": 1.1104,
+      "step": 6230
+    },
+    {
+      "epoch": 4.078431372549019,
+      "grad_norm": 0.9324519038200378,
+      "learning_rate": 0.0002,
+      "loss": 1.1538,
+      "step": 6240
+    },
+    {
+      "epoch": 4.084967320261438,
+      "grad_norm": 0.8723901510238647,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 6250
+    },
+    {
+      "epoch": 4.091503267973856,
+      "grad_norm": 0.8343415856361389,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 6260
+    },
+    {
+      "epoch": 4.098039215686274,
+      "grad_norm": 0.7490310072898865,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 6270
+    },
+    {
+      "epoch": 4.104575163398692,
+      "grad_norm": 0.8961182832717896,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 6280
+    },
+    {
+      "epoch": 4.111111111111111,
+      "grad_norm": 0.7124854922294617,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 6290
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 0.8338138461112976,
+      "learning_rate": 0.0002,
+      "loss": 1.0081,
+      "step": 6300
+    },
+    {
+      "epoch": 4.124183006535947,
+      "grad_norm": 0.8075833320617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 6310
+    },
+    {
+      "epoch": 4.130718954248366,
+      "grad_norm": 0.8069391846656799,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 6320
+    },
+    {
+      "epoch": 4.137254901960785,
+      "grad_norm": 0.9567893147468567,
+      "learning_rate": 0.0002,
+      "loss": 0.948,
+      "step": 6330
+    },
+    {
+      "epoch": 4.143790849673203,
+      "grad_norm": 1.2184662818908691,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 6340
+    },
+    {
+      "epoch": 4.150326797385621,
+      "grad_norm": 1.030976414680481,
+      "learning_rate": 0.0002,
+      "loss": 1.0756,
+      "step": 6350
+    },
+    {
+      "epoch": 4.1568627450980395,
+      "grad_norm": 0.9749957323074341,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6360
+    },
+    {
+      "epoch": 4.163398692810458,
+      "grad_norm": 0.7089483141899109,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 6370
+    },
+    {
+      "epoch": 4.169934640522876,
+      "grad_norm": 1.1084946393966675,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 6380
+    },
+    {
+      "epoch": 4.176470588235294,
+      "grad_norm": 0.7998497486114502,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 6390
+    },
+    {
+      "epoch": 4.183006535947713,
+      "grad_norm": 0.8997811675071716,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 6400
+    },
+    {
+      "epoch": 4.189542483660131,
+      "grad_norm": 0.8359479904174805,
+      "learning_rate": 0.0002,
+      "loss": 1.0704,
+      "step": 6410
+    },
+    {
+      "epoch": 4.196078431372549,
+      "grad_norm": 0.9087472558021545,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 6420
+    },
+    {
+      "epoch": 4.2026143790849675,
+      "grad_norm": 1.1100451946258545,
+      "learning_rate": 0.0002,
+      "loss": 1.0657,
+      "step": 6430
+    },
+    {
+      "epoch": 4.209150326797386,
+      "grad_norm": 0.9376999735832214,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 6440
+    },
+    {
+      "epoch": 4.215686274509804,
+      "grad_norm": 0.8179266452789307,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 6450
+    },
+    {
+      "epoch": 4.222222222222222,
+      "grad_norm": 0.9953271746635437,
+      "learning_rate": 0.0002,
+      "loss": 1.0679,
+      "step": 6460
+    },
+    {
+      "epoch": 4.228758169934641,
+      "grad_norm": 0.8476650714874268,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 6470
+    },
+    {
+      "epoch": 4.235294117647059,
+      "grad_norm": 0.8406323194503784,
+      "learning_rate": 0.0002,
+      "loss": 1.2512,
+      "step": 6480
+    },
+    {
+      "epoch": 4.241830065359477,
+      "grad_norm": 0.819134533405304,
+      "learning_rate": 0.0002,
+      "loss": 1.057,
+      "step": 6490
+    },
+    {
+      "epoch": 4.248366013071895,
+      "grad_norm": 0.7764983773231506,
+      "learning_rate": 0.0002,
+      "loss": 1.1082,
+      "step": 6500
+    },
+    {
+      "epoch": 4.254901960784314,
+      "grad_norm": 0.8252112865447998,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 6510
+    },
+    {
+      "epoch": 4.261437908496732,
+      "grad_norm": 0.7941019535064697,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 6520
+    },
+    {
+      "epoch": 4.26797385620915,
+      "grad_norm": 0.7673905491828918,
+      "learning_rate": 0.0002,
+      "loss": 1.0296,
+      "step": 6530
+    },
+    {
+      "epoch": 4.2745098039215685,
+      "grad_norm": 0.8749890327453613,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6540
+    },
+    {
+      "epoch": 4.281045751633987,
+      "grad_norm": 0.7343207597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 6550
+    },
+    {
+      "epoch": 4.287581699346405,
+      "grad_norm": 1.2786651849746704,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6560
+    },
+    {
+      "epoch": 4.294117647058823,
+      "grad_norm": 1.316875696182251,
+      "learning_rate": 0.0002,
+      "loss": 1.0514,
+      "step": 6570
+    },
+    {
+      "epoch": 4.300653594771242,
+      "grad_norm": 0.8349189162254333,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 6580
+    },
+    {
+      "epoch": 4.30718954248366,
+      "grad_norm": 0.7510647177696228,
+      "learning_rate": 0.0002,
+      "loss": 1.0732,
+      "step": 6590
+    },
+    {
+      "epoch": 4.313725490196078,
+      "grad_norm": 0.932420551776886,
+      "learning_rate": 0.0002,
+      "loss": 1.1387,
+      "step": 6600
+    },
+    {
+      "epoch": 4.3202614379084965,
+      "grad_norm": 0.8510616421699524,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 6610
+    },
+    {
+      "epoch": 4.326797385620915,
+      "grad_norm": 0.7661547064781189,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 6620
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 1.0370930433273315,
+      "learning_rate": 0.0002,
+      "loss": 1.2064,
+      "step": 6630
+    },
+    {
+      "epoch": 4.339869281045751,
+      "grad_norm": 0.9302158951759338,
+      "learning_rate": 0.0002,
+      "loss": 1.1064,
+      "step": 6640
+    },
+    {
+      "epoch": 4.34640522875817,
+      "grad_norm": 0.9203811883926392,
+      "learning_rate": 0.0002,
+      "loss": 0.968,
+      "step": 6650
+    },
+    {
+      "epoch": 4.352941176470588,
+      "grad_norm": 0.9986332654953003,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 6660
+    },
+    {
+      "epoch": 4.359477124183006,
+      "grad_norm": 0.8001713156700134,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 6670
+    },
+    {
+      "epoch": 4.366013071895424,
+      "grad_norm": 0.829714298248291,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 6680
+    },
+    {
+      "epoch": 4.372549019607844,
+      "grad_norm": 0.8253079056739807,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 6690
+    },
+    {
+      "epoch": 4.379084967320262,
+      "grad_norm": 0.824666440486908,
+      "learning_rate": 0.0002,
+      "loss": 1.1087,
+      "step": 6700
+    },
+    {
+      "epoch": 4.38562091503268,
+      "grad_norm": 0.8872972130775452,
+      "learning_rate": 0.0002,
+      "loss": 1.1968,
+      "step": 6710
+    },
+    {
+      "epoch": 4.392156862745098,
+      "grad_norm": 0.8729761838912964,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 6720
+    },
+    {
+      "epoch": 4.398692810457517,
+      "grad_norm": 1.1367264986038208,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 6730
+    },
+    {
+      "epoch": 4.405228758169935,
+      "grad_norm": 0.9699058532714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0184,
+      "step": 6740
+    },
+    {
+      "epoch": 4.411764705882353,
+      "grad_norm": 0.8266763687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.006,
+      "step": 6750
+    },
+    {
+      "epoch": 4.4183006535947715,
+      "grad_norm": 1.0249767303466797,
+      "learning_rate": 0.0002,
+      "loss": 1.0735,
+      "step": 6760
+    },
+    {
+      "epoch": 4.42483660130719,
+      "grad_norm": 0.73606938123703,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 6770
+    },
+    {
+      "epoch": 4.431372549019608,
+      "grad_norm": 1.4050679206848145,
+      "learning_rate": 0.0002,
+      "loss": 1.1037,
+      "step": 6780
+    },
+    {
+      "epoch": 4.437908496732026,
+      "grad_norm": 1.1114081144332886,
+      "learning_rate": 0.0002,
+      "loss": 1.1418,
+      "step": 6790
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.8031067848205566,
+      "learning_rate": 0.0002,
+      "loss": 0.9682,
+      "step": 6800
+    },
+    {
+      "epoch": 4.450980392156863,
+      "grad_norm": 0.8513566851615906,
+      "learning_rate": 0.0002,
+      "loss": 1.0753,
+      "step": 6810
+    },
+    {
+      "epoch": 4.457516339869281,
+      "grad_norm": 1.332741379737854,
+      "learning_rate": 0.0002,
+      "loss": 1.1852,
+      "step": 6820
+    },
+    {
+      "epoch": 4.4640522875816995,
+      "grad_norm": 1.5032578706741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 6830
+    },
+    {
+      "epoch": 4.470588235294118,
+      "grad_norm": 0.7677283883094788,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 6840
+    },
+    {
+      "epoch": 4.477124183006536,
+      "grad_norm": 0.989148736000061,
+      "learning_rate": 0.0002,
+      "loss": 1.1501,
+      "step": 6850
+    },
+    {
+      "epoch": 4.483660130718954,
+      "grad_norm": 1.5316275358200073,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 6860
+    },
+    {
+      "epoch": 4.490196078431373,
+      "grad_norm": 0.9427124261856079,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 6870
+    },
+    {
+      "epoch": 4.496732026143791,
+      "grad_norm": 1.215287685394287,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 6880
+    },
+    {
+      "epoch": 4.503267973856209,
+      "grad_norm": 0.7286760210990906,
+      "learning_rate": 0.0002,
+      "loss": 1.0809,
+      "step": 6890
+    },
+    {
+      "epoch": 4.509803921568627,
+      "grad_norm": 0.874829888343811,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 6900
+    },
+    {
+      "epoch": 4.516339869281046,
+      "grad_norm": 0.8058359622955322,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 6910
+    },
+    {
+      "epoch": 4.522875816993464,
+      "grad_norm": 1.248195767402649,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 6920
+    },
+    {
+      "epoch": 4.529411764705882,
+      "grad_norm": 0.8033645749092102,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 6930
+    },
+    {
+      "epoch": 4.5359477124183005,
+      "grad_norm": 1.7361950874328613,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 6940
+    },
+    {
+      "epoch": 4.542483660130719,
+      "grad_norm": 0.8058095574378967,
+      "learning_rate": 0.0002,
+      "loss": 0.9856,
+      "step": 6950
+    },
+    {
+      "epoch": 4.549019607843137,
+      "grad_norm": 1.254089593887329,
+      "learning_rate": 0.0002,
+      "loss": 1.0057,
+      "step": 6960
+    },
+    {
+      "epoch": 4.555555555555555,
+      "grad_norm": 0.9180455803871155,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 6970
+    },
+    {
+      "epoch": 4.562091503267974,
+      "grad_norm": 0.6677682399749756,
+      "learning_rate": 0.0002,
+      "loss": 1.0559,
+      "step": 6980
+    },
+    {
+      "epoch": 4.568627450980392,
+      "grad_norm": 0.8127354383468628,
+      "learning_rate": 0.0002,
+      "loss": 1.0453,
+      "step": 6990
+    },
+    {
+      "epoch": 4.57516339869281,
+      "grad_norm": 1.0263001918792725,
+      "learning_rate": 0.0002,
+      "loss": 1.0828,
+      "step": 7000
+    },
+    {
+      "epoch": 4.5816993464052285,
+      "grad_norm": 0.9641909003257751,
+      "learning_rate": 0.0002,
+      "loss": 1.0703,
+      "step": 7010
+    },
+    {
+      "epoch": 4.588235294117647,
+      "grad_norm": 0.9440861344337463,
+      "learning_rate": 0.0002,
+      "loss": 1.179,
+      "step": 7020
+    },
+    {
+      "epoch": 4.594771241830065,
+      "grad_norm": 0.9539011716842651,
+      "learning_rate": 0.0002,
+      "loss": 1.0931,
+      "step": 7030
+    },
+    {
+      "epoch": 4.601307189542483,
+      "grad_norm": 1.0449910163879395,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7040
+    },
+    {
+      "epoch": 4.607843137254902,
+      "grad_norm": 0.8766893744468689,
+      "learning_rate": 0.0002,
+      "loss": 0.9944,
+      "step": 7050
+    },
+    {
+      "epoch": 4.61437908496732,
+      "grad_norm": 0.6983462572097778,
+      "learning_rate": 0.0002,
+      "loss": 1.0169,
+      "step": 7060
+    },
+    {
+      "epoch": 4.620915032679738,
+      "grad_norm": 0.9505505561828613,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 7070
+    },
+    {
+      "epoch": 4.627450980392156,
+      "grad_norm": 1.2506657838821411,
+      "learning_rate": 0.0002,
+      "loss": 1.121,
+      "step": 7080
+    },
+    {
+      "epoch": 4.633986928104575,
+      "grad_norm": 0.9602801203727722,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 7090
+    },
+    {
+      "epoch": 4.640522875816993,
+      "grad_norm": 0.7398977875709534,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 7100
+    },
+    {
+      "epoch": 4.647058823529412,
+      "grad_norm": 1.3862425088882446,
+      "learning_rate": 0.0002,
+      "loss": 1.0769,
+      "step": 7110
+    },
+    {
+      "epoch": 4.65359477124183,
+      "grad_norm": 1.1451990604400635,
+      "learning_rate": 0.0002,
+      "loss": 1.0571,
+      "step": 7120
+    },
+    {
+      "epoch": 4.660130718954249,
+      "grad_norm": 0.9010422229766846,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 7130
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.7102518081665039,
+      "learning_rate": 0.0002,
+      "loss": 1.0165,
+      "step": 7140
+    },
+    {
+      "epoch": 4.673202614379085,
+      "grad_norm": 0.7963796257972717,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7150
+    },
+    {
+      "epoch": 4.6797385620915035,
+      "grad_norm": 0.7726007699966431,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 7160
+    },
+    {
+      "epoch": 4.686274509803922,
+      "grad_norm": 0.8097564578056335,
+      "learning_rate": 0.0002,
+      "loss": 1.2088,
+      "step": 7170
+    },
+    {
+      "epoch": 4.69281045751634,
+      "grad_norm": 0.9070925116539001,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 7180
+    },
+    {
+      "epoch": 4.699346405228758,
+      "grad_norm": 0.7543528079986572,
+      "learning_rate": 0.0002,
+      "loss": 1.0315,
+      "step": 7190
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.9900904893875122,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 7200
+    },
+    {
+      "epoch": 4.712418300653595,
+      "grad_norm": 0.8033412098884583,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 7210
+    },
+    {
+      "epoch": 4.718954248366013,
+      "grad_norm": 0.8440839052200317,
+      "learning_rate": 0.0002,
+      "loss": 1.1773,
+      "step": 7220
+    },
+    {
+      "epoch": 4.7254901960784315,
+      "grad_norm": 0.9325555562973022,
+      "learning_rate": 0.0002,
+      "loss": 1.1258,
+      "step": 7230
+    },
+    {
+      "epoch": 4.73202614379085,
+      "grad_norm": 0.7881146669387817,
+      "learning_rate": 0.0002,
+      "loss": 1.1384,
+      "step": 7240
+    },
+    {
+      "epoch": 4.738562091503268,
+      "grad_norm": 0.884453296661377,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7250
+    },
+    {
+      "epoch": 4.745098039215686,
+      "grad_norm": 0.9274539351463318,
+      "learning_rate": 0.0002,
+      "loss": 1.1036,
+      "step": 7260
+    },
+    {
+      "epoch": 4.751633986928105,
+      "grad_norm": 1.2367479801177979,
+      "learning_rate": 0.0002,
+      "loss": 1.0906,
+      "step": 7270
+    },
+    {
+      "epoch": 4.758169934640523,
+      "grad_norm": 0.9499821066856384,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 7280
+    },
+    {
+      "epoch": 4.764705882352941,
+      "grad_norm": 2.1918580532073975,
+      "learning_rate": 0.0002,
+      "loss": 1.1625,
+      "step": 7290
+    },
+    {
+      "epoch": 4.771241830065359,
+      "grad_norm": 0.8221880793571472,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 7300
+    },
+    {
+      "epoch": 4.777777777777778,
+      "grad_norm": 0.871972918510437,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7310
+    },
+    {
+      "epoch": 4.784313725490196,
+      "grad_norm": 0.8034510612487793,
+      "learning_rate": 0.0002,
+      "loss": 1.0599,
+      "step": 7320
+    },
+    {
+      "epoch": 4.790849673202614,
+      "grad_norm": 0.8959605693817139,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 7330
+    },
+    {
+      "epoch": 4.7973856209150325,
+      "grad_norm": 1.2326215505599976,
+      "learning_rate": 0.0002,
+      "loss": 1.0176,
+      "step": 7340
+    },
+    {
+      "epoch": 4.803921568627451,
+      "grad_norm": 0.9725791811943054,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7350
+    },
+    {
+      "epoch": 4.810457516339869,
+      "grad_norm": 0.7240816354751587,
+      "learning_rate": 0.0002,
+      "loss": 1.1229,
+      "step": 7360
+    },
+    {
+      "epoch": 4.816993464052287,
+      "grad_norm": 0.8265769481658936,
+      "learning_rate": 0.0002,
+      "loss": 1.0669,
+      "step": 7370
+    },
+    {
+      "epoch": 4.823529411764706,
+      "grad_norm": 0.8888696432113647,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 7380
+    },
+    {
+      "epoch": 4.830065359477124,
+      "grad_norm": 0.7776556015014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 7390
+    },
+    {
+      "epoch": 4.836601307189542,
+      "grad_norm": 0.8772371411323547,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7400
+    },
+    {
+      "epoch": 4.8431372549019605,
+      "grad_norm": 0.9786531925201416,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 7410
+    },
+    {
+      "epoch": 4.849673202614379,
+      "grad_norm": 0.9059745073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.1358,
+      "step": 7420
+    },
+    {
+      "epoch": 4.856209150326797,
+      "grad_norm": 0.7422552108764648,
+      "learning_rate": 0.0002,
+      "loss": 1.0324,
+      "step": 7430
+    },
+    {
+      "epoch": 4.862745098039216,
+      "grad_norm": 1.3040380477905273,
+      "learning_rate": 0.0002,
+      "loss": 1.0423,
+      "step": 7440
+    },
+    {
+      "epoch": 4.8692810457516345,
+      "grad_norm": 1.3278473615646362,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 7450
+    },
+    {
+      "epoch": 4.875816993464053,
+      "grad_norm": 1.2705849409103394,
+      "learning_rate": 0.0002,
+      "loss": 1.0713,
+      "step": 7460
+    },
+    {
+      "epoch": 4.882352941176471,
+      "grad_norm": 0.8837892413139343,
+      "learning_rate": 0.0002,
+      "loss": 1.0034,
+      "step": 7470
+    },
+    {
+      "epoch": 4.888888888888889,
+      "grad_norm": 0.8670691251754761,
+      "learning_rate": 0.0002,
+      "loss": 1.1716,
+      "step": 7480
+    },
+    {
+      "epoch": 4.895424836601308,
+      "grad_norm": 0.9662758111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.1723,
+      "step": 7490
+    },
+    {
+      "epoch": 4.901960784313726,
+      "grad_norm": 0.8188302516937256,
+      "learning_rate": 0.0002,
+      "loss": 1.1056,
+      "step": 7500
+    },
+    {
+      "epoch": 4.908496732026144,
+      "grad_norm": 0.769442617893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0419,
+      "step": 7510
+    },
+    {
+      "epoch": 4.915032679738562,
+      "grad_norm": 1.1465084552764893,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 7520
+    },
+    {
+      "epoch": 4.921568627450981,
+      "grad_norm": 1.253214955329895,
+      "learning_rate": 0.0002,
+      "loss": 1.0768,
+      "step": 7530
+    },
+    {
+      "epoch": 4.928104575163399,
+      "grad_norm": 0.7922375202178955,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 7540
+    },
+    {
+      "epoch": 4.934640522875817,
+      "grad_norm": 0.8306851387023926,
+      "learning_rate": 0.0002,
+      "loss": 1.1256,
+      "step": 7550
+    },
+    {
+      "epoch": 4.9411764705882355,
+      "grad_norm": 0.8486151099205017,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 7560
+    },
+    {
+      "epoch": 4.947712418300654,
+      "grad_norm": 1.2601467370986938,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7570
+    },
+    {
+      "epoch": 4.954248366013072,
+      "grad_norm": 0.7980747818946838,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 7580
+    },
+    {
+      "epoch": 4.96078431372549,
+      "grad_norm": 0.8653254508972168,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 7590
+    },
+    {
+      "epoch": 4.967320261437909,
+      "grad_norm": 0.9680571556091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 7600
+    },
+    {
+      "epoch": 4.973856209150327,
+      "grad_norm": 0.9554466605186462,
+      "learning_rate": 0.0002,
+      "loss": 1.1795,
+      "step": 7610
+    },
+    {
+      "epoch": 4.980392156862745,
+      "grad_norm": 1.3693897724151611,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 7620
+    },
+    {
+      "epoch": 4.9869281045751634,
+      "grad_norm": 0.7809282541275024,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 7630
+    },
+    {
+      "epoch": 4.993464052287582,
+      "grad_norm": 0.7528006434440613,
+      "learning_rate": 0.0002,
+      "loss": 1.0844,
+      "step": 7640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7491309642791748,
+      "learning_rate": 0.0002,
+      "loss": 0.9951,
+      "step": 7650
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.4197258949279785,
+      "eval_runtime": 33.6327,
+      "eval_samples_per_second": 12.964,
+      "eval_steps_per_second": 1.635,
+      "step": 7650
+    },
+    {
+      "epoch": 5.006535947712418,
+      "grad_norm": 0.8840063214302063,
+      "learning_rate": 0.0002,
+      "loss": 0.9744,
+      "step": 7660
+    },
+    {
+      "epoch": 5.0130718954248366,
+      "grad_norm": 1.0118401050567627,
+      "learning_rate": 0.0002,
+      "loss": 1.0274,
+      "step": 7670
+    },
+    {
+      "epoch": 5.019607843137255,
+      "grad_norm": 1.0040518045425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1667,
+      "step": 7680
+    },
+    {
+      "epoch": 5.026143790849673,
+      "grad_norm": 0.7541199922561646,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 7690
+    },
+    {
+      "epoch": 5.032679738562091,
+      "grad_norm": 0.9106482863426208,
+      "learning_rate": 0.0002,
+      "loss": 1.0797,
+      "step": 7700
+    },
+    {
+      "epoch": 5.03921568627451,
+      "grad_norm": 1.3691469430923462,
+      "learning_rate": 0.0002,
+      "loss": 1.0096,
+      "step": 7710
+    },
+    {
+      "epoch": 5.045751633986928,
+      "grad_norm": 0.9449689388275146,
+      "learning_rate": 0.0002,
+      "loss": 0.9889,
+      "step": 7720
+    },
+    {
+      "epoch": 5.052287581699346,
+      "grad_norm": 1.1678508520126343,
+      "learning_rate": 0.0002,
+      "loss": 0.9087,
+      "step": 7730
+    },
+    {
+      "epoch": 5.0588235294117645,
+      "grad_norm": 1.1296145915985107,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 7740
+    },
+    {
+      "epoch": 5.065359477124183,
+      "grad_norm": 0.7863904237747192,
+      "learning_rate": 0.0002,
+      "loss": 0.9339,
+      "step": 7750
+    },
+    {
+      "epoch": 5.071895424836601,
+      "grad_norm": 0.8691433072090149,
+      "learning_rate": 0.0002,
+      "loss": 1.0135,
+      "step": 7760
+    },
+    {
+      "epoch": 5.078431372549019,
+      "grad_norm": 1.0722088813781738,
+      "learning_rate": 0.0002,
+      "loss": 0.9776,
+      "step": 7770
+    },
+    {
+      "epoch": 5.084967320261438,
+      "grad_norm": 0.9625038504600525,
+      "learning_rate": 0.0002,
+      "loss": 1.0595,
+      "step": 7780
+    },
+    {
+      "epoch": 5.091503267973856,
+      "grad_norm": 1.2618783712387085,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 7790
+    },
+    {
+      "epoch": 5.098039215686274,
+      "grad_norm": 0.9970650672912598,
+      "learning_rate": 0.0002,
+      "loss": 0.9396,
+      "step": 7800
+    },
+    {
+      "epoch": 5.104575163398692,
+      "grad_norm": 1.3946677446365356,
+      "learning_rate": 0.0002,
+      "loss": 0.9186,
+      "step": 7810
+    },
+    {
+      "epoch": 5.111111111111111,
+      "grad_norm": 1.0260052680969238,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 7820
+    },
+    {
+      "epoch": 5.117647058823529,
+      "grad_norm": 1.105521559715271,
+      "learning_rate": 0.0002,
+      "loss": 0.9865,
+      "step": 7830
+    },
+    {
+      "epoch": 5.124183006535947,
+      "grad_norm": 1.003641128540039,
+      "learning_rate": 0.0002,
+      "loss": 0.9788,
+      "step": 7840
+    },
+    {
+      "epoch": 5.130718954248366,
+      "grad_norm": 1.0315021276474,
+      "learning_rate": 0.0002,
+      "loss": 0.9688,
+      "step": 7850
+    },
+    {
+      "epoch": 5.137254901960785,
+      "grad_norm": 0.9469530582427979,
+      "learning_rate": 0.0002,
+      "loss": 1.0001,
+      "step": 7860
+    },
+    {
+      "epoch": 5.143790849673203,
+      "grad_norm": 1.3244667053222656,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 7870
+    },
+    {
+      "epoch": 5.150326797385621,
+      "grad_norm": 1.1732033491134644,
+      "learning_rate": 0.0002,
+      "loss": 0.9657,
+      "step": 7880
+    },
+    {
+      "epoch": 5.1568627450980395,
+      "grad_norm": 1.3129149675369263,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 7890
+    },
+    {
+      "epoch": 5.163398692810458,
+      "grad_norm": 0.8589454293251038,
+      "learning_rate": 0.0002,
+      "loss": 0.9894,
+      "step": 7900
+    },
+    {
+      "epoch": 5.169934640522876,
+      "grad_norm": 0.8954233527183533,
+      "learning_rate": 0.0002,
+      "loss": 1.0161,
+      "step": 7910
+    },
+    {
+      "epoch": 5.176470588235294,
+      "grad_norm": 0.7426522970199585,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 7920
+    },
+    {
+      "epoch": 5.183006535947713,
+      "grad_norm": 1.1990121603012085,
+      "learning_rate": 0.0002,
+      "loss": 1.0106,
+      "step": 7930
+    },
+    {
+      "epoch": 5.189542483660131,
+      "grad_norm": 0.8867580890655518,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 7940
+    },
+    {
+      "epoch": 5.196078431372549,
+      "grad_norm": 1.016276478767395,
+      "learning_rate": 0.0002,
+      "loss": 0.9727,
+      "step": 7950
+    },
+    {
+      "epoch": 5.2026143790849675,
+      "grad_norm": 1.0210685729980469,
+      "learning_rate": 0.0002,
+      "loss": 0.9908,
+      "step": 7960
+    },
+    {
+      "epoch": 5.209150326797386,
+      "grad_norm": 1.0093122720718384,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 7970
+    },
+    {
+      "epoch": 5.215686274509804,
+      "grad_norm": 0.9746801853179932,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 7980
+    },
+    {
+      "epoch": 5.222222222222222,
+      "grad_norm": 0.9113537073135376,
+      "learning_rate": 0.0002,
+      "loss": 1.0611,
+      "step": 7990
+    },
+    {
+      "epoch": 5.228758169934641,
+      "grad_norm": 1.2782206535339355,
+      "learning_rate": 0.0002,
+      "loss": 0.9167,
+      "step": 8000
+    },
+    {
+      "epoch": 5.235294117647059,
+      "grad_norm": 1.3223118782043457,
+      "learning_rate": 0.0002,
+      "loss": 1.0212,
+      "step": 8010
+    },
+    {
+      "epoch": 5.241830065359477,
+      "grad_norm": 0.7898629307746887,
+      "learning_rate": 0.0002,
+      "loss": 0.9244,
+      "step": 8020
+    },
+    {
+      "epoch": 5.248366013071895,
+      "grad_norm": 0.9822350740432739,
+      "learning_rate": 0.0002,
+      "loss": 1.0574,
+      "step": 8030
+    },
+    {
+      "epoch": 5.254901960784314,
+      "grad_norm": 1.5114340782165527,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 8040
+    },
+    {
+      "epoch": 5.261437908496732,
+      "grad_norm": 0.859006941318512,
+      "learning_rate": 0.0002,
+      "loss": 0.9816,
+      "step": 8050
+    },
+    {
+      "epoch": 5.26797385620915,
+      "grad_norm": 1.0495043992996216,
+      "learning_rate": 0.0002,
+      "loss": 0.9445,
+      "step": 8060
+    },
+    {
+      "epoch": 5.2745098039215685,
+      "grad_norm": 1.329483151435852,
+      "learning_rate": 0.0002,
+      "loss": 0.9724,
+      "step": 8070
+    },
+    {
+      "epoch": 5.281045751633987,
+      "grad_norm": 1.1333061456680298,
+      "learning_rate": 0.0002,
+      "loss": 0.9296,
+      "step": 8080
+    },
+    {
+      "epoch": 5.287581699346405,
+      "grad_norm": 0.8153108358383179,
+      "learning_rate": 0.0002,
+      "loss": 0.9577,
+      "step": 8090
+    },
+    {
+      "epoch": 5.294117647058823,
+      "grad_norm": 0.9395004510879517,
+      "learning_rate": 0.0002,
+      "loss": 0.9002,
+      "step": 8100
+    },
+    {
+      "epoch": 5.300653594771242,
+      "grad_norm": 0.8907593488693237,
+      "learning_rate": 0.0002,
+      "loss": 1.0371,
+      "step": 8110
+    },
+    {
+      "epoch": 5.30718954248366,
+      "grad_norm": 0.9808667898178101,
+      "learning_rate": 0.0002,
+      "loss": 0.9301,
+      "step": 8120
+    },
+    {
+      "epoch": 5.313725490196078,
+      "grad_norm": 0.984779417514801,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 8130
+    },
+    {
+      "epoch": 5.3202614379084965,
+      "grad_norm": 0.9787270426750183,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 8140
+    },
+    {
+      "epoch": 5.326797385620915,
+      "grad_norm": 0.9857710599899292,
+      "learning_rate": 0.0002,
+      "loss": 0.9336,
+      "step": 8150
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9774303436279297,
+      "learning_rate": 0.0002,
+      "loss": 0.9884,
+      "step": 8160
+    },
+    {
+      "epoch": 5.339869281045751,
+      "grad_norm": 0.677925169467926,
+      "learning_rate": 0.0002,
+      "loss": 1.0561,
+      "step": 8170
+    },
+    {
+      "epoch": 5.34640522875817,
+      "grad_norm": 0.9576456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.1345,
+      "step": 8180
+    },
+    {
+      "epoch": 5.352941176470588,
+      "grad_norm": 1.8970937728881836,
+      "learning_rate": 0.0002,
+      "loss": 0.9554,
+      "step": 8190
+    },
+    {
+      "epoch": 5.359477124183006,
+      "grad_norm": 0.9458389282226562,
+      "learning_rate": 0.0002,
+      "loss": 1.0474,
+      "step": 8200
+    },
+    {
+      "epoch": 5.366013071895424,
+      "grad_norm": 1.761794924736023,
+      "learning_rate": 0.0002,
+      "loss": 1.0365,
+      "step": 8210
+    },
+    {
+      "epoch": 5.372549019607844,
+      "grad_norm": 1.0693724155426025,
+      "learning_rate": 0.0002,
+      "loss": 0.9426,
+      "step": 8220
+    },
+    {
+      "epoch": 5.379084967320262,
+      "grad_norm": 0.9025877714157104,
+      "learning_rate": 0.0002,
+      "loss": 1.0299,
+      "step": 8230
+    },
+    {
+      "epoch": 5.38562091503268,
+      "grad_norm": 1.258857250213623,
+      "learning_rate": 0.0002,
+      "loss": 0.9652,
+      "step": 8240
+    },
+    {
+      "epoch": 5.392156862745098,
+      "grad_norm": 1.084849238395691,
+      "learning_rate": 0.0002,
+      "loss": 0.9735,
+      "step": 8250
+    },
+    {
+      "epoch": 5.398692810457517,
+      "grad_norm": 0.9530340433120728,
+      "learning_rate": 0.0002,
+      "loss": 0.9999,
+      "step": 8260
+    },
+    {
+      "epoch": 5.405228758169935,
+      "grad_norm": 0.830240786075592,
+      "learning_rate": 0.0002,
+      "loss": 1.0268,
+      "step": 8270
+    },
+    {
+      "epoch": 5.411764705882353,
+      "grad_norm": 1.5807015895843506,
+      "learning_rate": 0.0002,
+      "loss": 1.0332,
+      "step": 8280
+    },
+    {
+      "epoch": 5.4183006535947715,
+      "grad_norm": 0.9486905336380005,
+      "learning_rate": 0.0002,
+      "loss": 0.9146,
+      "step": 8290
+    },
+    {
+      "epoch": 5.42483660130719,
+      "grad_norm": 1.0415093898773193,
+      "learning_rate": 0.0002,
+      "loss": 1.0336,
+      "step": 8300
+    },
+    {
+      "epoch": 5.431372549019608,
+      "grad_norm": 1.0501102209091187,
+      "learning_rate": 0.0002,
+      "loss": 0.8933,
+      "step": 8310
+    },
+    {
+      "epoch": 5.437908496732026,
+      "grad_norm": 0.9751836061477661,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 8320
+    },
+    {
+      "epoch": 5.444444444444445,
+      "grad_norm": 1.5529173612594604,
+      "learning_rate": 0.0002,
+      "loss": 1.0755,
+      "step": 8330
+    },
+    {
+      "epoch": 5.450980392156863,
+      "grad_norm": 0.8314350247383118,
+      "learning_rate": 0.0002,
+      "loss": 0.9814,
+      "step": 8340
+    },
+    {
+      "epoch": 5.457516339869281,
+      "grad_norm": 1.2555103302001953,
+      "learning_rate": 0.0002,
+      "loss": 1.0596,
+      "step": 8350
+    },
+    {
+      "epoch": 5.4640522875816995,
+      "grad_norm": 0.9408367872238159,
+      "learning_rate": 0.0002,
+      "loss": 1.0127,
+      "step": 8360
+    },
+    {
+      "epoch": 5.470588235294118,
+      "grad_norm": 0.9483312964439392,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 8370
+    },
+    {
+      "epoch": 5.477124183006536,
+      "grad_norm": 0.957905650138855,
+      "learning_rate": 0.0002,
+      "loss": 0.9678,
+      "step": 8380
+    },
+    {
+      "epoch": 5.483660130718954,
+      "grad_norm": 1.4000147581100464,
+      "learning_rate": 0.0002,
+      "loss": 1.0985,
+      "step": 8390
+    },
+    {
+      "epoch": 5.490196078431373,
+      "grad_norm": 1.7032461166381836,
+      "learning_rate": 0.0002,
+      "loss": 0.9966,
+      "step": 8400
+    },
+    {
+      "epoch": 5.496732026143791,
+      "grad_norm": 0.8978716731071472,
+      "learning_rate": 0.0002,
+      "loss": 0.9539,
+      "step": 8410
+    },
+    {
+      "epoch": 5.503267973856209,
+      "grad_norm": 0.8659300804138184,
+      "learning_rate": 0.0002,
+      "loss": 0.9544,
+      "step": 8420
+    },
+    {
+      "epoch": 5.509803921568627,
+      "grad_norm": 1.3629727363586426,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 8430
+    },
+    {
+      "epoch": 5.516339869281046,
+      "grad_norm": 1.2741984128952026,
+      "learning_rate": 0.0002,
+      "loss": 0.9696,
+      "step": 8440
+    },
+    {
+      "epoch": 5.522875816993464,
+      "grad_norm": 1.3867180347442627,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 8450
+    },
+    {
+      "epoch": 5.529411764705882,
+      "grad_norm": 1.0662001371383667,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 8460
+    },
+    {
+      "epoch": 5.5359477124183005,
+      "grad_norm": 1.7005380392074585,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 8470
+    },
+    {
+      "epoch": 5.542483660130719,
+      "grad_norm": 1.3730385303497314,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 8480
+    },
+    {
+      "epoch": 5.549019607843137,
+      "grad_norm": 1.7737441062927246,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 8490
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.907487690448761,
+      "learning_rate": 0.0002,
+      "loss": 0.9729,
+      "step": 8500
+    },
+    {
+      "epoch": 5.562091503267974,
+      "grad_norm": 0.8882441520690918,
+      "learning_rate": 0.0002,
+      "loss": 0.9891,
+      "step": 8510
+    },
+    {
+      "epoch": 5.568627450980392,
+      "grad_norm": 0.8655388951301575,
+      "learning_rate": 0.0002,
+      "loss": 0.973,
+      "step": 8520
+    },
+    {
+      "epoch": 5.57516339869281,
+      "grad_norm": 1.379992961883545,
+      "learning_rate": 0.0002,
+      "loss": 0.9523,
+      "step": 8530
+    },
+    {
+      "epoch": 5.5816993464052285,
+      "grad_norm": 1.0021201372146606,
+      "learning_rate": 0.0002,
+      "loss": 1.0174,
+      "step": 8540
+    },
+    {
+      "epoch": 5.588235294117647,
+      "grad_norm": 1.2636926174163818,
+      "learning_rate": 0.0002,
+      "loss": 1.0113,
+      "step": 8550
+    },
+    {
+      "epoch": 5.594771241830065,
+      "grad_norm": 1.279025912284851,
+      "learning_rate": 0.0002,
+      "loss": 1.0243,
+      "step": 8560
+    },
+    {
+      "epoch": 5.601307189542483,
+      "grad_norm": 0.8885834217071533,
+      "learning_rate": 0.0002,
+      "loss": 0.9917,
+      "step": 8570
+    },
+    {
+      "epoch": 5.607843137254902,
+      "grad_norm": 1.1975032091140747,
+      "learning_rate": 0.0002,
+      "loss": 0.9849,
+      "step": 8580
+    },
+    {
+      "epoch": 5.61437908496732,
+      "grad_norm": 1.005470871925354,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8590
+    },
+    {
+      "epoch": 5.620915032679738,
+      "grad_norm": 1.104286551475525,
+      "learning_rate": 0.0002,
+      "loss": 0.9947,
+      "step": 8600
+    },
+    {
+      "epoch": 5.627450980392156,
+      "grad_norm": 1.435445785522461,
+      "learning_rate": 0.0002,
+      "loss": 1.0585,
+      "step": 8610
+    },
+    {
+      "epoch": 5.633986928104575,
+      "grad_norm": 1.0270172357559204,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 8620
+    },
+    {
+      "epoch": 5.640522875816993,
+      "grad_norm": 1.0929527282714844,
+      "learning_rate": 0.0002,
+      "loss": 1.0522,
+      "step": 8630
+    },
+    {
+      "epoch": 5.647058823529412,
+      "grad_norm": 1.1061221361160278,
+      "learning_rate": 0.0002,
+      "loss": 0.9694,
+      "step": 8640
+    },
+    {
+      "epoch": 5.65359477124183,
+      "grad_norm": 0.9563149213790894,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 8650
+    },
+    {
+      "epoch": 5.660130718954249,
+      "grad_norm": 1.0434954166412354,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 8660
+    },
+    {
+      "epoch": 5.666666666666667,
+      "grad_norm": 1.3695117235183716,
+      "learning_rate": 0.0002,
+      "loss": 0.9463,
+      "step": 8670
+    },
+    {
+      "epoch": 5.673202614379085,
+      "grad_norm": 1.0540564060211182,
+      "learning_rate": 0.0002,
+      "loss": 0.9441,
+      "step": 8680
+    },
+    {
+      "epoch": 5.6797385620915035,
+      "grad_norm": 1.5942492485046387,
+      "learning_rate": 0.0002,
+      "loss": 0.9755,
+      "step": 8690
+    },
+    {
+      "epoch": 5.686274509803922,
+      "grad_norm": 0.9485495090484619,
+      "learning_rate": 0.0002,
+      "loss": 1.0071,
+      "step": 8700
+    },
+    {
+      "epoch": 5.69281045751634,
+      "grad_norm": 1.1483162641525269,
+      "learning_rate": 0.0002,
+      "loss": 0.9998,
+      "step": 8710
+    },
+    {
+      "epoch": 5.699346405228758,
+      "grad_norm": 0.9075471758842468,
+      "learning_rate": 0.0002,
+      "loss": 0.9578,
+      "step": 8720
+    },
+    {
+      "epoch": 5.705882352941177,
+      "grad_norm": 1.7908551692962646,
+      "learning_rate": 0.0002,
+      "loss": 0.9488,
+      "step": 8730
+    },
+    {
+      "epoch": 5.712418300653595,
+      "grad_norm": 0.8867162466049194,
+      "learning_rate": 0.0002,
+      "loss": 1.0163,
+      "step": 8740
+    },
+    {
+      "epoch": 5.718954248366013,
+      "grad_norm": 1.7165148258209229,
+      "learning_rate": 0.0002,
+      "loss": 1.0041,
+      "step": 8750
+    },
+    {
+      "epoch": 5.7254901960784315,
+      "grad_norm": 0.9529356956481934,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 8760
+    },
+    {
+      "epoch": 5.73202614379085,
+      "grad_norm": 1.01852548122406,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 8770
+    },
+    {
+      "epoch": 5.738562091503268,
+      "grad_norm": 0.9538423418998718,
+      "learning_rate": 0.0002,
+      "loss": 1.0471,
+      "step": 8780
+    },
+    {
+      "epoch": 5.745098039215686,
+      "grad_norm": 0.9007737636566162,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 8790
+    },
+    {
+      "epoch": 5.751633986928105,
+      "grad_norm": 0.9107874035835266,
+      "learning_rate": 0.0002,
+      "loss": 0.9766,
+      "step": 8800
+    },
+    {
+      "epoch": 5.758169934640523,
+      "grad_norm": 0.7379238605499268,
+      "learning_rate": 0.0002,
+      "loss": 0.9212,
+      "step": 8810
+    },
+    {
+      "epoch": 5.764705882352941,
+      "grad_norm": 1.072645902633667,
+      "learning_rate": 0.0002,
+      "loss": 1.0966,
+      "step": 8820
+    },
+    {
+      "epoch": 5.771241830065359,
+      "grad_norm": 1.002008080482483,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 8830
+    },
+    {
+      "epoch": 5.777777777777778,
+      "grad_norm": 1.0435924530029297,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 8840
+    },
+    {
+      "epoch": 5.784313725490196,
+      "grad_norm": 0.9874551296234131,
+      "learning_rate": 0.0002,
+      "loss": 0.9458,
+      "step": 8850
+    },
+    {
+      "epoch": 5.790849673202614,
+      "grad_norm": 1.1729662418365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1241,
+      "step": 8860
+    },
+    {
+      "epoch": 5.7973856209150325,
+      "grad_norm": 1.3300775289535522,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 8870
+    },
+    {
+      "epoch": 5.803921568627451,
+      "grad_norm": 1.612707257270813,
+      "learning_rate": 0.0002,
+      "loss": 1.0989,
+      "step": 8880
+    },
+    {
+      "epoch": 5.810457516339869,
+      "grad_norm": 0.9047797322273254,
+      "learning_rate": 0.0002,
+      "loss": 0.9119,
+      "step": 8890
+    },
+    {
+      "epoch": 5.816993464052287,
+      "grad_norm": 1.0958741903305054,
+      "learning_rate": 0.0002,
+      "loss": 0.989,
+      "step": 8900
+    },
+    {
+      "epoch": 5.823529411764706,
+      "grad_norm": 1.0099612474441528,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 8910
+    },
+    {
+      "epoch": 5.830065359477124,
+      "grad_norm": 0.8442328572273254,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 8920
+    },
+    {
+      "epoch": 5.836601307189542,
+      "grad_norm": 1.1388301849365234,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 8930
+    },
+    {
+      "epoch": 5.8431372549019605,
+      "grad_norm": 0.8296026587486267,
+      "learning_rate": 0.0002,
+      "loss": 1.0019,
+      "step": 8940
+    },
+    {
+      "epoch": 5.849673202614379,
+      "grad_norm": 1.0843533277511597,
+      "learning_rate": 0.0002,
+      "loss": 1.0363,
+      "step": 8950
+    },
+    {
+      "epoch": 5.856209150326797,
+      "grad_norm": 0.8496834635734558,
+      "learning_rate": 0.0002,
+      "loss": 1.0009,
+      "step": 8960
+    },
+    {
+      "epoch": 5.862745098039216,
+      "grad_norm": 1.6894690990447998,
+      "learning_rate": 0.0002,
+      "loss": 0.9927,
+      "step": 8970
+    },
+    {
+      "epoch": 5.8692810457516345,
+      "grad_norm": 1.0012282133102417,
+      "learning_rate": 0.0002,
+      "loss": 1.0939,
+      "step": 8980
+    },
+    {
+      "epoch": 5.875816993464053,
+      "grad_norm": 0.8521103262901306,
+      "learning_rate": 0.0002,
+      "loss": 0.9722,
+      "step": 8990
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 1.246841311454773,
+      "learning_rate": 0.0002,
+      "loss": 1.0885,
+      "step": 9000
+    },
+    {
+      "epoch": 5.888888888888889,
+      "grad_norm": 0.9941892027854919,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 9010
+    },
+    {
+      "epoch": 5.895424836601308,
+      "grad_norm": 1.067413568496704,
+      "learning_rate": 0.0002,
+      "loss": 0.8754,
+      "step": 9020
+    },
+    {
+      "epoch": 5.901960784313726,
+      "grad_norm": 1.0045088529586792,
+      "learning_rate": 0.0002,
+      "loss": 1.0153,
+      "step": 9030
+    },
+    {
+      "epoch": 5.908496732026144,
+      "grad_norm": 1.383063554763794,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 9040
+    },
+    {
+      "epoch": 5.915032679738562,
+      "grad_norm": 0.8754428625106812,
+      "learning_rate": 0.0002,
+      "loss": 1.0845,
+      "step": 9050
+    },
+    {
+      "epoch": 5.921568627450981,
+      "grad_norm": 0.8577388525009155,
+      "learning_rate": 0.0002,
+      "loss": 0.9571,
+      "step": 9060
+    },
+    {
+      "epoch": 5.928104575163399,
+      "grad_norm": 0.8718975186347961,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 9070
+    },
+    {
+      "epoch": 5.934640522875817,
+      "grad_norm": 1.1762131452560425,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 9080
+    },
+    {
+      "epoch": 5.9411764705882355,
+      "grad_norm": 1.1025866270065308,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 9090
+    },
+    {
+      "epoch": 5.947712418300654,
+      "grad_norm": 1.0439870357513428,
+      "learning_rate": 0.0002,
+      "loss": 0.9155,
+      "step": 9100
+    },
+    {
+      "epoch": 5.954248366013072,
+      "grad_norm": 1.2411525249481201,
+      "learning_rate": 0.0002,
+      "loss": 1.0055,
+      "step": 9110
+    },
+    {
+      "epoch": 5.96078431372549,
+      "grad_norm": 1.0317714214324951,
+      "learning_rate": 0.0002,
+      "loss": 0.9747,
+      "step": 9120
+    },
+    {
+      "epoch": 5.967320261437909,
+      "grad_norm": 0.9880492091178894,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 9130
+    },
+    {
+      "epoch": 5.973856209150327,
+      "grad_norm": 0.9039815664291382,
+      "learning_rate": 0.0002,
+      "loss": 1.0459,
+      "step": 9140
+    },
+    {
+      "epoch": 5.980392156862745,
+      "grad_norm": 0.9049116373062134,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 9150
+    },
+    {
+      "epoch": 5.9869281045751634,
+      "grad_norm": 0.996749222278595,
+      "learning_rate": 0.0002,
+      "loss": 0.9792,
+      "step": 9160
+    },
+    {
+      "epoch": 5.993464052287582,
+      "grad_norm": 0.8716062307357788,
+      "learning_rate": 0.0002,
+      "loss": 0.8857,
+      "step": 9170
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.3081294298171997,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 9180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.45111083984375,
+      "eval_runtime": 34.7121,
+      "eval_samples_per_second": 12.56,
+      "eval_steps_per_second": 1.584,
+      "step": 9180
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 12240,
+  "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": 1.1219792635625472e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-9180/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/special_tokens_map.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..f58963a682665634ab180c28667e4faa8cf02ba2
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f559f2189f392b4555613965f089e7c4d300b41fbe080bf79da0d676e33ee7f0
+size 34356041
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.model b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer_config.json b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1adb4796c13b8d975555ecec45876ee75d1ae8b7
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/tokenizer_config.json
@@ -0,0 +1,1757 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_args.bin b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..35e722282419bcef977427e4d3675fe3b94ec688
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25f531ee37172f22a819ab79094fe89aae41504e4c8b696743b5e23d9e7641
+size 5560
diff --git a/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_log.jsonl b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..9a6dd0a9b7b5310236d26ee5defe5aabea697f6b
--- /dev/null
+++ b/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 1.0, "step": 1530, "epoch_duration": 1647.488445520401, "total_accumulated_duration": 1647.488445520401, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 5628.7490234375}, "avg_memory_reserved": {"GPU_0": 6182.0}, "peak_memory_reserved": {"GPU_0": 6182.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}]}
+{"epoch": 2.0, "step": 3060, "epoch_duration": 1650.3675973415375, "total_accumulated_duration": 3297.8560428619385, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-1530", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}]}
+{"epoch": 3.0, "step": 4590, "epoch_duration": 1615.1267035007477, "total_accumulated_duration": 4912.982746362686, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-3060", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}]}
+{"epoch": 4.0, "step": 6120, "epoch_duration": 1762.4439125061035, "total_accumulated_duration": 6675.42665886879, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-4590", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}, {"eval_loss": 1.4131312370300293, "eval_runtime": 31.8967, "eval_samples_per_second": 13.669, "eval_steps_per_second": 1.724, "epoch": 3.0, "step": 4590}, {"loss": 1.1283, "grad_norm": 0.6926420331001282, "learning_rate": 0.0002, "epoch": 3.0065359477124183, "step": 4600}, {"loss": 1.1537, "grad_norm": 0.6656355857849121, "learning_rate": 0.0002, "epoch": 3.0130718954248366, "step": 4610}, {"loss": 1.308, "grad_norm": 0.9901936650276184, "learning_rate": 0.0002, "epoch": 3.019607843137255, "step": 4620}, {"loss": 1.22, "grad_norm": 0.6713474988937378, "learning_rate": 0.0002, "epoch": 3.026143790849673, "step": 4630}, {"loss": 1.2249, "grad_norm": 0.6199324131011963, "learning_rate": 0.0002, "epoch": 3.0326797385620914, "step": 4640}, {"loss": 1.242, "grad_norm": 0.7180785536766052, "learning_rate": 0.0002, "epoch": 3.0392156862745097, "step": 4650}, {"loss": 1.1349, "grad_norm": 0.8256588578224182, "learning_rate": 0.0002, "epoch": 3.045751633986928, "step": 4660}, {"loss": 1.1431, "grad_norm": 0.6637389063835144, "learning_rate": 0.0002, "epoch": 3.052287581699346, "step": 4670}, {"loss": 1.1096, "grad_norm": 0.6980698108673096, "learning_rate": 0.0002, "epoch": 3.0588235294117645, "step": 4680}, {"loss": 1.196, "grad_norm": 0.8091534972190857, "learning_rate": 0.0002, "epoch": 3.065359477124183, "step": 4690}, {"loss": 1.1652, "grad_norm": 0.5715174078941345, "learning_rate": 0.0002, "epoch": 3.0718954248366015, "step": 4700}, {"loss": 1.1427, "grad_norm": 0.735639750957489, "learning_rate": 0.0002, "epoch": 3.0784313725490198, "step": 4710}, {"loss": 1.1522, "grad_norm": 0.7619708180427551, "learning_rate": 0.0002, "epoch": 3.084967320261438, "step": 4720}, {"loss": 1.0853, "grad_norm": 1.263566017150879, "learning_rate": 0.0002, "epoch": 3.0915032679738563, "step": 4730}, {"loss": 1.1348, "grad_norm": 0.6600871682167053, "learning_rate": 0.0002, "epoch": 3.0980392156862746, "step": 4740}, {"loss": 1.1766, "grad_norm": 0.717792809009552, "learning_rate": 0.0002, "epoch": 3.104575163398693, "step": 4750}, {"loss": 1.088, "grad_norm": 0.853714644908905, "learning_rate": 0.0002, "epoch": 3.111111111111111, "step": 4760}, {"loss": 1.2031, "grad_norm": 1.1004153490066528, "learning_rate": 0.0002, "epoch": 3.1176470588235294, "step": 4770}, {"loss": 1.3295, "grad_norm": 0.8566235899925232, "learning_rate": 0.0002, "epoch": 3.1241830065359477, "step": 4780}, {"loss": 1.2436, "grad_norm": 0.8315296173095703, "learning_rate": 0.0002, "epoch": 3.130718954248366, "step": 4790}, {"loss": 1.32, "grad_norm": 0.8020524978637695, "learning_rate": 0.0002, "epoch": 3.1372549019607843, "step": 4800}, {"loss": 1.1238, "grad_norm": 0.7564275860786438, "learning_rate": 0.0002, "epoch": 3.1437908496732025, "step": 4810}, {"loss": 1.1244, "grad_norm": 0.9077776670455933, "learning_rate": 0.0002, "epoch": 3.150326797385621, "step": 4820}, {"loss": 1.1399, "grad_norm": 0.6323099732398987, "learning_rate": 0.0002, "epoch": 3.156862745098039, "step": 4830}, {"loss": 1.1983, "grad_norm": 0.6625368595123291, "learning_rate": 0.0002, "epoch": 3.1633986928104574, "step": 4840}, {"loss": 1.066, "grad_norm": 0.8119261860847473, "learning_rate": 0.0002, "epoch": 3.1699346405228757, "step": 4850}, {"loss": 1.0224, "grad_norm": 0.6399450898170471, "learning_rate": 0.0002, "epoch": 3.176470588235294, "step": 4860}, {"loss": 1.2181, "grad_norm": 1.0659016370773315, "learning_rate": 0.0002, "epoch": 3.183006535947712, "step": 4870}, {"loss": 1.2914, "grad_norm": 0.8040369749069214, "learning_rate": 0.0002, "epoch": 3.189542483660131, "step": 4880}, {"loss": 1.1996, "grad_norm": 0.7784733176231384, "learning_rate": 0.0002, "epoch": 3.196078431372549, "step": 4890}, {"loss": 1.2051, "grad_norm": 0.9660294651985168, "learning_rate": 0.0002, "epoch": 3.2026143790849675, "step": 4900}, {"loss": 1.0419, "grad_norm": 1.0676977634429932, "learning_rate": 0.0002, "epoch": 3.2091503267973858, "step": 4910}, {"loss": 1.0083, "grad_norm": 0.5877565741539001, "learning_rate": 0.0002, "epoch": 3.215686274509804, "step": 4920}, {"loss": 1.1046, "grad_norm": 0.6164032816886902, "learning_rate": 0.0002, "epoch": 3.2222222222222223, "step": 4930}, {"loss": 1.1079, "grad_norm": 0.7627606987953186, "learning_rate": 0.0002, "epoch": 3.2287581699346406, "step": 4940}, {"loss": 1.2453, "grad_norm": 0.7442803978919983, "learning_rate": 0.0002, "epoch": 3.235294117647059, "step": 4950}, {"loss": 1.1087, "grad_norm": 0.7277812361717224, "learning_rate": 0.0002, "epoch": 3.241830065359477, "step": 4960}, {"loss": 1.2237, "grad_norm": 1.0301902294158936, "learning_rate": 0.0002, "epoch": 3.2483660130718954, "step": 4970}, {"loss": 1.1466, "grad_norm": 0.7798232436180115, "learning_rate": 0.0002, "epoch": 3.2549019607843137, "step": 4980}, {"loss": 1.2142, "grad_norm": 1.210265874862671, "learning_rate": 0.0002, "epoch": 3.261437908496732, "step": 4990}, {"loss": 1.1557, "grad_norm": 0.6677713990211487, "learning_rate": 0.0002, "epoch": 3.2679738562091503, "step": 5000}, {"loss": 1.3294, "grad_norm": 1.0524500608444214, "learning_rate": 0.0002, "epoch": 3.2745098039215685, "step": 5010}, {"loss": 1.1939, "grad_norm": 0.7091745734214783, "learning_rate": 0.0002, "epoch": 3.281045751633987, "step": 5020}, {"loss": 1.1891, "grad_norm": 0.8523224592208862, "learning_rate": 0.0002, "epoch": 3.287581699346405, "step": 5030}, {"loss": 1.1925, "grad_norm": 0.6120608448982239, "learning_rate": 0.0002, "epoch": 3.2941176470588234, "step": 5040}, {"loss": 1.0603, "grad_norm": 0.7437472939491272, "learning_rate": 0.0002, "epoch": 3.3006535947712417, "step": 5050}, {"loss": 1.1295, "grad_norm": 0.7611715197563171, "learning_rate": 0.0002, "epoch": 3.30718954248366, "step": 5060}, {"loss": 1.0531, "grad_norm": 0.7249704003334045, "learning_rate": 0.0002, "epoch": 3.313725490196078, "step": 5070}, {"loss": 1.2292, "grad_norm": 0.7316247820854187, "learning_rate": 0.0002, "epoch": 3.3202614379084965, "step": 5080}, {"loss": 1.1974, "grad_norm": 0.562412440776825, "learning_rate": 0.0002, "epoch": 3.326797385620915, "step": 5090}, {"loss": 1.0736, "grad_norm": 0.7052176594734192, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 5100}, {"loss": 1.122, "grad_norm": 0.7714211344718933, "learning_rate": 0.0002, "epoch": 3.3398692810457518, "step": 5110}, {"loss": 1.1684, "grad_norm": 1.0436055660247803, "learning_rate": 0.0002, "epoch": 3.34640522875817, "step": 5120}, {"loss": 1.0945, "grad_norm": 0.8867271542549133, "learning_rate": 0.0002, "epoch": 3.3529411764705883, "step": 5130}, {"loss": 1.159, "grad_norm": 0.8371267914772034, "learning_rate": 0.0002, "epoch": 3.3594771241830066, "step": 5140}, {"loss": 1.1073, "grad_norm": 0.7257837057113647, "learning_rate": 0.0002, "epoch": 3.366013071895425, "step": 5150}, {"loss": 1.1162, "grad_norm": 0.7102002501487732, "learning_rate": 0.0002, "epoch": 3.372549019607843, "step": 5160}, {"loss": 1.2056, "grad_norm": 0.7636350393295288, "learning_rate": 0.0002, "epoch": 3.3790849673202614, "step": 5170}, {"loss": 1.0708, "grad_norm": 0.6887359619140625, "learning_rate": 0.0002, "epoch": 3.3856209150326797, "step": 5180}, {"loss": 1.3807, "grad_norm": 0.8141424655914307, "learning_rate": 0.0002, "epoch": 3.392156862745098, "step": 5190}, {"loss": 1.1986, "grad_norm": 0.694423496723175, "learning_rate": 0.0002, "epoch": 3.3986928104575163, "step": 5200}, {"loss": 1.2945, "grad_norm": 0.914013683795929, "learning_rate": 0.0002, "epoch": 3.4052287581699345, "step": 5210}, {"loss": 1.1413, "grad_norm": 0.8503239750862122, "learning_rate": 0.0002, "epoch": 3.411764705882353, "step": 5220}, {"loss": 1.2696, "grad_norm": 0.6196836233139038, "learning_rate": 0.0002, "epoch": 3.418300653594771, "step": 5230}, {"loss": 1.2431, "grad_norm": 1.0760811567306519, "learning_rate": 0.0002, "epoch": 3.4248366013071894, "step": 5240}, {"loss": 1.1686, "grad_norm": 0.6524698138237, "learning_rate": 0.0002, "epoch": 3.431372549019608, "step": 5250}, {"loss": 1.2012, "grad_norm": 0.674467921257019, "learning_rate": 0.0002, "epoch": 3.4379084967320264, "step": 5260}, {"loss": 1.1015, "grad_norm": 0.7690372467041016, "learning_rate": 0.0002, "epoch": 3.4444444444444446, "step": 5270}, {"loss": 1.2511, "grad_norm": 0.8751813769340515, "learning_rate": 0.0002, "epoch": 3.450980392156863, "step": 5280}, {"loss": 1.1841, "grad_norm": 0.750407874584198, "learning_rate": 0.0002, "epoch": 3.457516339869281, "step": 5290}, {"loss": 1.0605, "grad_norm": 0.5991823077201843, "learning_rate": 0.0002, "epoch": 3.4640522875816995, "step": 5300}, {"loss": 1.2347, "grad_norm": 1.0164772272109985, "learning_rate": 0.0002, "epoch": 3.4705882352941178, "step": 5310}, {"loss": 1.2354, "grad_norm": 0.8704105019569397, "learning_rate": 0.0002, "epoch": 3.477124183006536, "step": 5320}, {"loss": 1.2169, "grad_norm": 0.709102213382721, "learning_rate": 0.0002, "epoch": 3.4836601307189543, "step": 5330}, {"loss": 1.2425, "grad_norm": 0.6273632049560547, "learning_rate": 0.0002, "epoch": 3.4901960784313726, "step": 5340}, {"loss": 1.1585, "grad_norm": 0.6807359457015991, "learning_rate": 0.0002, "epoch": 3.496732026143791, "step": 5350}, {"loss": 1.131, "grad_norm": 0.7085188627243042, "learning_rate": 0.0002, "epoch": 3.503267973856209, "step": 5360}, {"loss": 1.1159, "grad_norm": 0.6938307881355286, "learning_rate": 0.0002, "epoch": 3.5098039215686274, "step": 5370}, {"loss": 1.1397, "grad_norm": 0.8544146418571472, "learning_rate": 0.0002, "epoch": 3.5163398692810457, "step": 5380}, {"loss": 1.2181, "grad_norm": 0.7889642119407654, "learning_rate": 0.0002, "epoch": 3.522875816993464, "step": 5390}, {"loss": 1.1691, "grad_norm": 0.7858421206474304, "learning_rate": 0.0002, "epoch": 3.5294117647058822, "step": 5400}, {"loss": 1.2374, "grad_norm": 0.8547123074531555, "learning_rate": 0.0002, "epoch": 3.5359477124183005, "step": 5410}, {"loss": 1.196, "grad_norm": 0.8218181133270264, "learning_rate": 0.0002, "epoch": 3.542483660130719, "step": 5420}, {"loss": 1.1961, "grad_norm": 1.153623342514038, "learning_rate": 0.0002, "epoch": 3.549019607843137, "step": 5430}, {"loss": 1.156, "grad_norm": 1.1321099996566772, "learning_rate": 0.0002, "epoch": 3.5555555555555554, "step": 5440}, {"loss": 1.2224, "grad_norm": 0.9495334029197693, "learning_rate": 0.0002, "epoch": 3.5620915032679736, "step": 5450}, {"loss": 1.2869, "grad_norm": 0.8743821978569031, "learning_rate": 0.0002, "epoch": 3.568627450980392, "step": 5460}, {"loss": 1.1018, "grad_norm": 0.7513086795806885, "learning_rate": 0.0002, "epoch": 3.57516339869281, "step": 5470}, {"loss": 1.1082, "grad_norm": 1.0139480829238892, "learning_rate": 0.0002, "epoch": 3.581699346405229, "step": 5480}, {"loss": 1.1706, "grad_norm": 0.6615135073661804, "learning_rate": 0.0002, "epoch": 3.588235294117647, "step": 5490}, {"loss": 1.3906, "grad_norm": 1.180798888206482, "learning_rate": 0.0002, "epoch": 3.5947712418300655, "step": 5500}, {"loss": 1.2391, "grad_norm": 0.7085279226303101, "learning_rate": 0.0002, "epoch": 3.6013071895424837, "step": 5510}, {"loss": 1.1623, "grad_norm": 0.540268063545227, "learning_rate": 0.0002, "epoch": 3.607843137254902, "step": 5520}, {"loss": 1.2132, "grad_norm": 0.7905671000480652, "learning_rate": 0.0002, "epoch": 3.6143790849673203, "step": 5530}, {"loss": 1.2731, "grad_norm": 0.8457717299461365, "learning_rate": 0.0002, "epoch": 3.6209150326797386, "step": 5540}, {"loss": 1.1799, "grad_norm": 0.7102677822113037, "learning_rate": 0.0002, "epoch": 3.627450980392157, "step": 5550}, {"loss": 1.2394, "grad_norm": 0.7179514765739441, "learning_rate": 0.0002, "epoch": 3.633986928104575, "step": 5560}, {"loss": 1.2019, "grad_norm": 1.0854148864746094, "learning_rate": 0.0002, "epoch": 3.6405228758169934, "step": 5570}, {"loss": 1.1986, "grad_norm": 0.8209951519966125, "learning_rate": 0.0002, "epoch": 3.6470588235294117, "step": 5580}, {"loss": 1.2289, "grad_norm": 0.6944138407707214, "learning_rate": 0.0002, "epoch": 3.65359477124183, "step": 5590}, {"loss": 1.3226, "grad_norm": 0.7675473093986511, "learning_rate": 0.0002, "epoch": 3.6601307189542482, "step": 5600}, {"loss": 1.2866, "grad_norm": 0.6683364510536194, "learning_rate": 0.0002, "epoch": 3.6666666666666665, "step": 5610}, {"loss": 1.1099, "grad_norm": 0.7920727133750916, "learning_rate": 0.0002, "epoch": 3.6732026143790852, "step": 5620}, {"loss": 1.2287, "grad_norm": 0.9440218806266785, "learning_rate": 0.0002, "epoch": 3.6797385620915035, "step": 5630}, {"loss": 1.2444, "grad_norm": 0.6600824594497681, "learning_rate": 0.0002, "epoch": 3.686274509803922, "step": 5640}, {"loss": 1.191, "grad_norm": 0.6860619187355042, "learning_rate": 0.0002, "epoch": 3.69281045751634, "step": 5650}, {"loss": 1.1914, "grad_norm": 0.6579713225364685, "learning_rate": 0.0002, "epoch": 3.6993464052287583, "step": 5660}, {"loss": 1.1464, "grad_norm": 0.661081075668335, "learning_rate": 0.0002, "epoch": 3.7058823529411766, "step": 5670}, {"loss": 1.289, "grad_norm": 1.0968825817108154, "learning_rate": 0.0002, "epoch": 3.712418300653595, "step": 5680}, {"loss": 1.192, "grad_norm": 0.8066844940185547, "learning_rate": 0.0002, "epoch": 3.718954248366013, "step": 5690}, {"loss": 1.2322, "grad_norm": 0.8341682553291321, "learning_rate": 0.0002, "epoch": 3.7254901960784315, "step": 5700}, {"loss": 1.1473, "grad_norm": 0.6682852506637573, "learning_rate": 0.0002, "epoch": 3.7320261437908497, "step": 5710}, {"loss": 1.1566, "grad_norm": 0.898595929145813, "learning_rate": 0.0002, "epoch": 3.738562091503268, "step": 5720}, {"loss": 1.0919, "grad_norm": 0.6876054406166077, "learning_rate": 0.0002, "epoch": 3.7450980392156863, "step": 5730}, {"loss": 1.2302, "grad_norm": 0.7817103266716003, "learning_rate": 0.0002, "epoch": 3.7516339869281046, "step": 5740}, {"loss": 1.2439, "grad_norm": 0.5840168595314026, "learning_rate": 0.0002, "epoch": 3.758169934640523, "step": 5750}, {"loss": 1.1279, "grad_norm": 0.6263918876647949, "learning_rate": 0.0002, "epoch": 3.764705882352941, "step": 5760}, {"loss": 1.2023, "grad_norm": 0.7948952317237854, "learning_rate": 0.0002, "epoch": 3.7712418300653594, "step": 5770}, {"loss": 1.149, "grad_norm": 0.6700998544692993, "learning_rate": 0.0002, "epoch": 3.7777777777777777, "step": 5780}, {"loss": 1.3207, "grad_norm": 1.1169519424438477, "learning_rate": 0.0002, "epoch": 3.784313725490196, "step": 5790}, {"loss": 1.064, "grad_norm": 0.8354471325874329, "learning_rate": 0.0002, "epoch": 3.7908496732026142, "step": 5800}, {"loss": 1.2104, "grad_norm": 0.6304181814193726, "learning_rate": 0.0002, "epoch": 3.7973856209150325, "step": 5810}, {"loss": 1.2059, "grad_norm": 0.6919655799865723, "learning_rate": 0.0002, "epoch": 3.803921568627451, "step": 5820}, {"loss": 1.217, "grad_norm": 0.600385844707489, "learning_rate": 0.0002, "epoch": 3.810457516339869, "step": 5830}, {"loss": 1.2324, "grad_norm": 0.8406319618225098, "learning_rate": 0.0002, "epoch": 3.8169934640522873, "step": 5840}, {"loss": 1.2418, "grad_norm": 0.7594282031059265, "learning_rate": 0.0002, "epoch": 3.8235294117647056, "step": 5850}, {"loss": 1.1903, "grad_norm": 0.8179879784584045, "learning_rate": 0.0002, "epoch": 3.8300653594771243, "step": 5860}, {"loss": 1.255, "grad_norm": 1.141430377960205, "learning_rate": 0.0002, "epoch": 3.8366013071895426, "step": 5870}, {"loss": 1.1467, "grad_norm": 0.6595550775527954, "learning_rate": 0.0002, "epoch": 3.843137254901961, "step": 5880}, {"loss": 1.2378, "grad_norm": 0.7499435544013977, "learning_rate": 0.0002, "epoch": 3.849673202614379, "step": 5890}, {"loss": 1.217, "grad_norm": 0.7851517200469971, "learning_rate": 0.0002, "epoch": 3.8562091503267975, "step": 5900}, {"loss": 1.162, "grad_norm": 1.0533545017242432, "learning_rate": 0.0002, "epoch": 3.8627450980392157, "step": 5910}, {"loss": 1.3576, "grad_norm": 0.960086464881897, "learning_rate": 0.0002, "epoch": 3.869281045751634, "step": 5920}, {"loss": 1.151, "grad_norm": 0.9952049851417542, "learning_rate": 0.0002, "epoch": 3.8758169934640523, "step": 5930}, {"loss": 1.2027, "grad_norm": 0.7884191274642944, "learning_rate": 0.0002, "epoch": 3.8823529411764706, "step": 5940}, {"loss": 1.1796, "grad_norm": 0.7461766600608826, "learning_rate": 0.0002, "epoch": 3.888888888888889, "step": 5950}, {"loss": 1.2251, "grad_norm": 0.9594355821609497, "learning_rate": 0.0002, "epoch": 3.895424836601307, "step": 5960}, {"loss": 1.1164, "grad_norm": 0.8179471492767334, "learning_rate": 0.0002, "epoch": 3.9019607843137254, "step": 5970}, {"loss": 1.2421, "grad_norm": 0.8240267634391785, "learning_rate": 0.0002, "epoch": 3.9084967320261437, "step": 5980}, {"loss": 1.3076, "grad_norm": 0.7462618350982666, "learning_rate": 0.0002, "epoch": 3.915032679738562, "step": 5990}, {"loss": 1.2124, "grad_norm": 0.711207389831543, "learning_rate": 0.0002, "epoch": 3.9215686274509802, "step": 6000}, {"loss": 1.2119, "grad_norm": 0.6910956501960754, "learning_rate": 0.0002, "epoch": 3.928104575163399, "step": 6010}, {"loss": 1.2127, "grad_norm": 0.749093770980835, "learning_rate": 0.0002, "epoch": 3.9346405228758172, "step": 6020}, {"loss": 1.1542, "grad_norm": 1.3332762718200684, "learning_rate": 0.0002, "epoch": 3.9411764705882355, "step": 6030}, {"loss": 1.1442, "grad_norm": 0.71457439661026, "learning_rate": 0.0002, "epoch": 3.947712418300654, "step": 6040}, {"loss": 1.339, "grad_norm": 1.1205238103866577, "learning_rate": 0.0002, "epoch": 3.954248366013072, "step": 6050}, {"loss": 1.2962, "grad_norm": 0.6958928108215332, "learning_rate": 0.0002, "epoch": 3.9607843137254903, "step": 6060}, {"loss": 1.1802, "grad_norm": 0.7518056035041809, "learning_rate": 0.0002, "epoch": 3.9673202614379086, "step": 6070}, {"loss": 1.1179, "grad_norm": 0.8010755777359009, "learning_rate": 0.0002, "epoch": 3.973856209150327, "step": 6080}, {"loss": 1.2867, "grad_norm": 0.7492658495903015, "learning_rate": 0.0002, "epoch": 3.980392156862745, "step": 6090}, {"loss": 1.2113, "grad_norm": 0.900704562664032, "learning_rate": 0.0002, "epoch": 3.9869281045751634, "step": 6100}, {"loss": 1.1106, "grad_norm": 0.7997331619262695, "learning_rate": 0.0002, "epoch": 3.9934640522875817, "step": 6110}, {"loss": 1.1244, "grad_norm": 0.7163209319114685, "learning_rate": 0.0002, "epoch": 4.0, "step": 6120}]}
+{"epoch": 5.0, "step": 7650, "epoch_duration": 1809.9564287662506, "total_accumulated_duration": 8485.38308763504, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}, {"eval_loss": 1.4131312370300293, "eval_runtime": 31.8967, "eval_samples_per_second": 13.669, "eval_steps_per_second": 1.724, "epoch": 3.0, "step": 4590}, {"loss": 1.1283, "grad_norm": 0.6926420331001282, "learning_rate": 0.0002, "epoch": 3.0065359477124183, "step": 4600}, {"loss": 1.1537, "grad_norm": 0.6656355857849121, "learning_rate": 0.0002, "epoch": 3.0130718954248366, "step": 4610}, {"loss": 1.308, "grad_norm": 0.9901936650276184, "learning_rate": 0.0002, "epoch": 3.019607843137255, "step": 4620}, {"loss": 1.22, "grad_norm": 0.6713474988937378, "learning_rate": 0.0002, "epoch": 3.026143790849673, "step": 4630}, {"loss": 1.2249, "grad_norm": 0.6199324131011963, "learning_rate": 0.0002, "epoch": 3.0326797385620914, "step": 4640}, {"loss": 1.242, "grad_norm": 0.7180785536766052, "learning_rate": 0.0002, "epoch": 3.0392156862745097, "step": 4650}, {"loss": 1.1349, "grad_norm": 0.8256588578224182, "learning_rate": 0.0002, "epoch": 3.045751633986928, "step": 4660}, {"loss": 1.1431, "grad_norm": 0.6637389063835144, "learning_rate": 0.0002, "epoch": 3.052287581699346, "step": 4670}, {"loss": 1.1096, "grad_norm": 0.6980698108673096, "learning_rate": 0.0002, "epoch": 3.0588235294117645, "step": 4680}, {"loss": 1.196, "grad_norm": 0.8091534972190857, "learning_rate": 0.0002, "epoch": 3.065359477124183, "step": 4690}, {"loss": 1.1652, "grad_norm": 0.5715174078941345, "learning_rate": 0.0002, "epoch": 3.0718954248366015, "step": 4700}, {"loss": 1.1427, "grad_norm": 0.735639750957489, "learning_rate": 0.0002, "epoch": 3.0784313725490198, "step": 4710}, {"loss": 1.1522, "grad_norm": 0.7619708180427551, "learning_rate": 0.0002, "epoch": 3.084967320261438, "step": 4720}, {"loss": 1.0853, "grad_norm": 1.263566017150879, "learning_rate": 0.0002, "epoch": 3.0915032679738563, "step": 4730}, {"loss": 1.1348, "grad_norm": 0.6600871682167053, "learning_rate": 0.0002, "epoch": 3.0980392156862746, "step": 4740}, {"loss": 1.1766, "grad_norm": 0.717792809009552, "learning_rate": 0.0002, "epoch": 3.104575163398693, "step": 4750}, {"loss": 1.088, "grad_norm": 0.853714644908905, "learning_rate": 0.0002, "epoch": 3.111111111111111, "step": 4760}, {"loss": 1.2031, "grad_norm": 1.1004153490066528, "learning_rate": 0.0002, "epoch": 3.1176470588235294, "step": 4770}, {"loss": 1.3295, "grad_norm": 0.8566235899925232, "learning_rate": 0.0002, "epoch": 3.1241830065359477, "step": 4780}, {"loss": 1.2436, "grad_norm": 0.8315296173095703, "learning_rate": 0.0002, "epoch": 3.130718954248366, "step": 4790}, {"loss": 1.32, "grad_norm": 0.8020524978637695, "learning_rate": 0.0002, "epoch": 3.1372549019607843, "step": 4800}, {"loss": 1.1238, "grad_norm": 0.7564275860786438, "learning_rate": 0.0002, "epoch": 3.1437908496732025, "step": 4810}, {"loss": 1.1244, "grad_norm": 0.9077776670455933, "learning_rate": 0.0002, "epoch": 3.150326797385621, "step": 4820}, {"loss": 1.1399, "grad_norm": 0.6323099732398987, "learning_rate": 0.0002, "epoch": 3.156862745098039, "step": 4830}, {"loss": 1.1983, "grad_norm": 0.6625368595123291, "learning_rate": 0.0002, "epoch": 3.1633986928104574, "step": 4840}, {"loss": 1.066, "grad_norm": 0.8119261860847473, "learning_rate": 0.0002, "epoch": 3.1699346405228757, "step": 4850}, {"loss": 1.0224, "grad_norm": 0.6399450898170471, "learning_rate": 0.0002, "epoch": 3.176470588235294, "step": 4860}, {"loss": 1.2181, "grad_norm": 1.0659016370773315, "learning_rate": 0.0002, "epoch": 3.183006535947712, "step": 4870}, {"loss": 1.2914, "grad_norm": 0.8040369749069214, "learning_rate": 0.0002, "epoch": 3.189542483660131, "step": 4880}, {"loss": 1.1996, "grad_norm": 0.7784733176231384, "learning_rate": 0.0002, "epoch": 3.196078431372549, "step": 4890}, {"loss": 1.2051, "grad_norm": 0.9660294651985168, "learning_rate": 0.0002, "epoch": 3.2026143790849675, "step": 4900}, {"loss": 1.0419, "grad_norm": 1.0676977634429932, "learning_rate": 0.0002, "epoch": 3.2091503267973858, "step": 4910}, {"loss": 1.0083, "grad_norm": 0.5877565741539001, "learning_rate": 0.0002, "epoch": 3.215686274509804, "step": 4920}, {"loss": 1.1046, "grad_norm": 0.6164032816886902, "learning_rate": 0.0002, "epoch": 3.2222222222222223, "step": 4930}, {"loss": 1.1079, "grad_norm": 0.7627606987953186, "learning_rate": 0.0002, "epoch": 3.2287581699346406, "step": 4940}, {"loss": 1.2453, "grad_norm": 0.7442803978919983, "learning_rate": 0.0002, "epoch": 3.235294117647059, "step": 4950}, {"loss": 1.1087, "grad_norm": 0.7277812361717224, "learning_rate": 0.0002, "epoch": 3.241830065359477, "step": 4960}, {"loss": 1.2237, "grad_norm": 1.0301902294158936, "learning_rate": 0.0002, "epoch": 3.2483660130718954, "step": 4970}, {"loss": 1.1466, "grad_norm": 0.7798232436180115, "learning_rate": 0.0002, "epoch": 3.2549019607843137, "step": 4980}, {"loss": 1.2142, "grad_norm": 1.210265874862671, "learning_rate": 0.0002, "epoch": 3.261437908496732, "step": 4990}, {"loss": 1.1557, "grad_norm": 0.6677713990211487, "learning_rate": 0.0002, "epoch": 3.2679738562091503, "step": 5000}, {"loss": 1.3294, "grad_norm": 1.0524500608444214, "learning_rate": 0.0002, "epoch": 3.2745098039215685, "step": 5010}, {"loss": 1.1939, "grad_norm": 0.7091745734214783, "learning_rate": 0.0002, "epoch": 3.281045751633987, "step": 5020}, {"loss": 1.1891, "grad_norm": 0.8523224592208862, "learning_rate": 0.0002, "epoch": 3.287581699346405, "step": 5030}, {"loss": 1.1925, "grad_norm": 0.6120608448982239, "learning_rate": 0.0002, "epoch": 3.2941176470588234, "step": 5040}, {"loss": 1.0603, "grad_norm": 0.7437472939491272, "learning_rate": 0.0002, "epoch": 3.3006535947712417, "step": 5050}, {"loss": 1.1295, "grad_norm": 0.7611715197563171, "learning_rate": 0.0002, "epoch": 3.30718954248366, "step": 5060}, {"loss": 1.0531, "grad_norm": 0.7249704003334045, "learning_rate": 0.0002, "epoch": 3.313725490196078, "step": 5070}, {"loss": 1.2292, "grad_norm": 0.7316247820854187, "learning_rate": 0.0002, "epoch": 3.3202614379084965, "step": 5080}, {"loss": 1.1974, "grad_norm": 0.562412440776825, "learning_rate": 0.0002, "epoch": 3.326797385620915, "step": 5090}, {"loss": 1.0736, "grad_norm": 0.7052176594734192, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 5100}, {"loss": 1.122, "grad_norm": 0.7714211344718933, "learning_rate": 0.0002, "epoch": 3.3398692810457518, "step": 5110}, {"loss": 1.1684, "grad_norm": 1.0436055660247803, "learning_rate": 0.0002, "epoch": 3.34640522875817, "step": 5120}, {"loss": 1.0945, "grad_norm": 0.8867271542549133, "learning_rate": 0.0002, "epoch": 3.3529411764705883, "step": 5130}, {"loss": 1.159, "grad_norm": 0.8371267914772034, "learning_rate": 0.0002, "epoch": 3.3594771241830066, "step": 5140}, {"loss": 1.1073, "grad_norm": 0.7257837057113647, "learning_rate": 0.0002, "epoch": 3.366013071895425, "step": 5150}, {"loss": 1.1162, "grad_norm": 0.7102002501487732, "learning_rate": 0.0002, "epoch": 3.372549019607843, "step": 5160}, {"loss": 1.2056, "grad_norm": 0.7636350393295288, "learning_rate": 0.0002, "epoch": 3.3790849673202614, "step": 5170}, {"loss": 1.0708, "grad_norm": 0.6887359619140625, "learning_rate": 0.0002, "epoch": 3.3856209150326797, "step": 5180}, {"loss": 1.3807, "grad_norm": 0.8141424655914307, "learning_rate": 0.0002, "epoch": 3.392156862745098, "step": 5190}, {"loss": 1.1986, "grad_norm": 0.694423496723175, "learning_rate": 0.0002, "epoch": 3.3986928104575163, "step": 5200}, {"loss": 1.2945, "grad_norm": 0.914013683795929, "learning_rate": 0.0002, "epoch": 3.4052287581699345, "step": 5210}, {"loss": 1.1413, "grad_norm": 0.8503239750862122, "learning_rate": 0.0002, "epoch": 3.411764705882353, "step": 5220}, {"loss": 1.2696, "grad_norm": 0.6196836233139038, "learning_rate": 0.0002, "epoch": 3.418300653594771, "step": 5230}, {"loss": 1.2431, "grad_norm": 1.0760811567306519, "learning_rate": 0.0002, "epoch": 3.4248366013071894, "step": 5240}, {"loss": 1.1686, "grad_norm": 0.6524698138237, "learning_rate": 0.0002, "epoch": 3.431372549019608, "step": 5250}, {"loss": 1.2012, "grad_norm": 0.674467921257019, "learning_rate": 0.0002, "epoch": 3.4379084967320264, "step": 5260}, {"loss": 1.1015, "grad_norm": 0.7690372467041016, "learning_rate": 0.0002, "epoch": 3.4444444444444446, "step": 5270}, {"loss": 1.2511, "grad_norm": 0.8751813769340515, "learning_rate": 0.0002, "epoch": 3.450980392156863, "step": 5280}, {"loss": 1.1841, "grad_norm": 0.750407874584198, "learning_rate": 0.0002, "epoch": 3.457516339869281, "step": 5290}, {"loss": 1.0605, "grad_norm": 0.5991823077201843, "learning_rate": 0.0002, "epoch": 3.4640522875816995, "step": 5300}, {"loss": 1.2347, "grad_norm": 1.0164772272109985, "learning_rate": 0.0002, "epoch": 3.4705882352941178, "step": 5310}, {"loss": 1.2354, "grad_norm": 0.8704105019569397, "learning_rate": 0.0002, "epoch": 3.477124183006536, "step": 5320}, {"loss": 1.2169, "grad_norm": 0.709102213382721, "learning_rate": 0.0002, "epoch": 3.4836601307189543, "step": 5330}, {"loss": 1.2425, "grad_norm": 0.6273632049560547, "learning_rate": 0.0002, "epoch": 3.4901960784313726, "step": 5340}, {"loss": 1.1585, "grad_norm": 0.6807359457015991, "learning_rate": 0.0002, "epoch": 3.496732026143791, "step": 5350}, {"loss": 1.131, "grad_norm": 0.7085188627243042, "learning_rate": 0.0002, "epoch": 3.503267973856209, "step": 5360}, {"loss": 1.1159, "grad_norm": 0.6938307881355286, "learning_rate": 0.0002, "epoch": 3.5098039215686274, "step": 5370}, {"loss": 1.1397, "grad_norm": 0.8544146418571472, "learning_rate": 0.0002, "epoch": 3.5163398692810457, "step": 5380}, {"loss": 1.2181, "grad_norm": 0.7889642119407654, "learning_rate": 0.0002, "epoch": 3.522875816993464, "step": 5390}, {"loss": 1.1691, "grad_norm": 0.7858421206474304, "learning_rate": 0.0002, "epoch": 3.5294117647058822, "step": 5400}, {"loss": 1.2374, "grad_norm": 0.8547123074531555, "learning_rate": 0.0002, "epoch": 3.5359477124183005, "step": 5410}, {"loss": 1.196, "grad_norm": 0.8218181133270264, "learning_rate": 0.0002, "epoch": 3.542483660130719, "step": 5420}, {"loss": 1.1961, "grad_norm": 1.153623342514038, "learning_rate": 0.0002, "epoch": 3.549019607843137, "step": 5430}, {"loss": 1.156, "grad_norm": 1.1321099996566772, "learning_rate": 0.0002, "epoch": 3.5555555555555554, "step": 5440}, {"loss": 1.2224, "grad_norm": 0.9495334029197693, "learning_rate": 0.0002, "epoch": 3.5620915032679736, "step": 5450}, {"loss": 1.2869, "grad_norm": 0.8743821978569031, "learning_rate": 0.0002, "epoch": 3.568627450980392, "step": 5460}, {"loss": 1.1018, "grad_norm": 0.7513086795806885, "learning_rate": 0.0002, "epoch": 3.57516339869281, "step": 5470}, {"loss": 1.1082, "grad_norm": 1.0139480829238892, "learning_rate": 0.0002, "epoch": 3.581699346405229, "step": 5480}, {"loss": 1.1706, "grad_norm": 0.6615135073661804, "learning_rate": 0.0002, "epoch": 3.588235294117647, "step": 5490}, {"loss": 1.3906, "grad_norm": 1.180798888206482, "learning_rate": 0.0002, "epoch": 3.5947712418300655, "step": 5500}, {"loss": 1.2391, "grad_norm": 0.7085279226303101, "learning_rate": 0.0002, "epoch": 3.6013071895424837, "step": 5510}, {"loss": 1.1623, "grad_norm": 0.540268063545227, "learning_rate": 0.0002, "epoch": 3.607843137254902, "step": 5520}, {"loss": 1.2132, "grad_norm": 0.7905671000480652, "learning_rate": 0.0002, "epoch": 3.6143790849673203, "step": 5530}, {"loss": 1.2731, "grad_norm": 0.8457717299461365, "learning_rate": 0.0002, "epoch": 3.6209150326797386, "step": 5540}, {"loss": 1.1799, "grad_norm": 0.7102677822113037, "learning_rate": 0.0002, "epoch": 3.627450980392157, "step": 5550}, {"loss": 1.2394, "grad_norm": 0.7179514765739441, "learning_rate": 0.0002, "epoch": 3.633986928104575, "step": 5560}, {"loss": 1.2019, "grad_norm": 1.0854148864746094, "learning_rate": 0.0002, "epoch": 3.6405228758169934, "step": 5570}, {"loss": 1.1986, "grad_norm": 0.8209951519966125, "learning_rate": 0.0002, "epoch": 3.6470588235294117, "step": 5580}, {"loss": 1.2289, "grad_norm": 0.6944138407707214, "learning_rate": 0.0002, "epoch": 3.65359477124183, "step": 5590}, {"loss": 1.3226, "grad_norm": 0.7675473093986511, "learning_rate": 0.0002, "epoch": 3.6601307189542482, "step": 5600}, {"loss": 1.2866, "grad_norm": 0.6683364510536194, "learning_rate": 0.0002, "epoch": 3.6666666666666665, "step": 5610}, {"loss": 1.1099, "grad_norm": 0.7920727133750916, "learning_rate": 0.0002, "epoch": 3.6732026143790852, "step": 5620}, {"loss": 1.2287, "grad_norm": 0.9440218806266785, "learning_rate": 0.0002, "epoch": 3.6797385620915035, "step": 5630}, {"loss": 1.2444, "grad_norm": 0.6600824594497681, "learning_rate": 0.0002, "epoch": 3.686274509803922, "step": 5640}, {"loss": 1.191, "grad_norm": 0.6860619187355042, "learning_rate": 0.0002, "epoch": 3.69281045751634, "step": 5650}, {"loss": 1.1914, "grad_norm": 0.6579713225364685, "learning_rate": 0.0002, "epoch": 3.6993464052287583, "step": 5660}, {"loss": 1.1464, "grad_norm": 0.661081075668335, "learning_rate": 0.0002, "epoch": 3.7058823529411766, "step": 5670}, {"loss": 1.289, "grad_norm": 1.0968825817108154, "learning_rate": 0.0002, "epoch": 3.712418300653595, "step": 5680}, {"loss": 1.192, "grad_norm": 0.8066844940185547, "learning_rate": 0.0002, "epoch": 3.718954248366013, "step": 5690}, {"loss": 1.2322, "grad_norm": 0.8341682553291321, "learning_rate": 0.0002, "epoch": 3.7254901960784315, "step": 5700}, {"loss": 1.1473, "grad_norm": 0.6682852506637573, "learning_rate": 0.0002, "epoch": 3.7320261437908497, "step": 5710}, {"loss": 1.1566, "grad_norm": 0.898595929145813, "learning_rate": 0.0002, "epoch": 3.738562091503268, "step": 5720}, {"loss": 1.0919, "grad_norm": 0.6876054406166077, "learning_rate": 0.0002, "epoch": 3.7450980392156863, "step": 5730}, {"loss": 1.2302, "grad_norm": 0.7817103266716003, "learning_rate": 0.0002, "epoch": 3.7516339869281046, "step": 5740}, {"loss": 1.2439, "grad_norm": 0.5840168595314026, "learning_rate": 0.0002, "epoch": 3.758169934640523, "step": 5750}, {"loss": 1.1279, "grad_norm": 0.6263918876647949, "learning_rate": 0.0002, "epoch": 3.764705882352941, "step": 5760}, {"loss": 1.2023, "grad_norm": 0.7948952317237854, "learning_rate": 0.0002, "epoch": 3.7712418300653594, "step": 5770}, {"loss": 1.149, "grad_norm": 0.6700998544692993, "learning_rate": 0.0002, "epoch": 3.7777777777777777, "step": 5780}, {"loss": 1.3207, "grad_norm": 1.1169519424438477, "learning_rate": 0.0002, "epoch": 3.784313725490196, "step": 5790}, {"loss": 1.064, "grad_norm": 0.8354471325874329, "learning_rate": 0.0002, "epoch": 3.7908496732026142, "step": 5800}, {"loss": 1.2104, "grad_norm": 0.6304181814193726, "learning_rate": 0.0002, "epoch": 3.7973856209150325, "step": 5810}, {"loss": 1.2059, "grad_norm": 0.6919655799865723, "learning_rate": 0.0002, "epoch": 3.803921568627451, "step": 5820}, {"loss": 1.217, "grad_norm": 0.600385844707489, "learning_rate": 0.0002, "epoch": 3.810457516339869, "step": 5830}, {"loss": 1.2324, "grad_norm": 0.8406319618225098, "learning_rate": 0.0002, "epoch": 3.8169934640522873, "step": 5840}, {"loss": 1.2418, "grad_norm": 0.7594282031059265, "learning_rate": 0.0002, "epoch": 3.8235294117647056, "step": 5850}, {"loss": 1.1903, "grad_norm": 0.8179879784584045, "learning_rate": 0.0002, "epoch": 3.8300653594771243, "step": 5860}, {"loss": 1.255, "grad_norm": 1.141430377960205, "learning_rate": 0.0002, "epoch": 3.8366013071895426, "step": 5870}, {"loss": 1.1467, "grad_norm": 0.6595550775527954, "learning_rate": 0.0002, "epoch": 3.843137254901961, "step": 5880}, {"loss": 1.2378, "grad_norm": 0.7499435544013977, "learning_rate": 0.0002, "epoch": 3.849673202614379, "step": 5890}, {"loss": 1.217, "grad_norm": 0.7851517200469971, "learning_rate": 0.0002, "epoch": 3.8562091503267975, "step": 5900}, {"loss": 1.162, "grad_norm": 1.0533545017242432, "learning_rate": 0.0002, "epoch": 3.8627450980392157, "step": 5910}, {"loss": 1.3576, "grad_norm": 0.960086464881897, "learning_rate": 0.0002, "epoch": 3.869281045751634, "step": 5920}, {"loss": 1.151, "grad_norm": 0.9952049851417542, "learning_rate": 0.0002, "epoch": 3.8758169934640523, "step": 5930}, {"loss": 1.2027, "grad_norm": 0.7884191274642944, "learning_rate": 0.0002, "epoch": 3.8823529411764706, "step": 5940}, {"loss": 1.1796, "grad_norm": 0.7461766600608826, "learning_rate": 0.0002, "epoch": 3.888888888888889, "step": 5950}, {"loss": 1.2251, "grad_norm": 0.9594355821609497, "learning_rate": 0.0002, "epoch": 3.895424836601307, "step": 5960}, {"loss": 1.1164, "grad_norm": 0.8179471492767334, "learning_rate": 0.0002, "epoch": 3.9019607843137254, "step": 5970}, {"loss": 1.2421, "grad_norm": 0.8240267634391785, "learning_rate": 0.0002, "epoch": 3.9084967320261437, "step": 5980}, {"loss": 1.3076, "grad_norm": 0.7462618350982666, "learning_rate": 0.0002, "epoch": 3.915032679738562, "step": 5990}, {"loss": 1.2124, "grad_norm": 0.711207389831543, "learning_rate": 0.0002, "epoch": 3.9215686274509802, "step": 6000}, {"loss": 1.2119, "grad_norm": 0.6910956501960754, "learning_rate": 0.0002, "epoch": 3.928104575163399, "step": 6010}, {"loss": 1.2127, "grad_norm": 0.749093770980835, "learning_rate": 0.0002, "epoch": 3.9346405228758172, "step": 6020}, {"loss": 1.1542, "grad_norm": 1.3332762718200684, "learning_rate": 0.0002, "epoch": 3.9411764705882355, "step": 6030}, {"loss": 1.1442, "grad_norm": 0.71457439661026, "learning_rate": 0.0002, "epoch": 3.947712418300654, "step": 6040}, {"loss": 1.339, "grad_norm": 1.1205238103866577, "learning_rate": 0.0002, "epoch": 3.954248366013072, "step": 6050}, {"loss": 1.2962, "grad_norm": 0.6958928108215332, "learning_rate": 0.0002, "epoch": 3.9607843137254903, "step": 6060}, {"loss": 1.1802, "grad_norm": 0.7518056035041809, "learning_rate": 0.0002, "epoch": 3.9673202614379086, "step": 6070}, {"loss": 1.1179, "grad_norm": 0.8010755777359009, "learning_rate": 0.0002, "epoch": 3.973856209150327, "step": 6080}, {"loss": 1.2867, "grad_norm": 0.7492658495903015, "learning_rate": 0.0002, "epoch": 3.980392156862745, "step": 6090}, {"loss": 1.2113, "grad_norm": 0.900704562664032, "learning_rate": 0.0002, "epoch": 3.9869281045751634, "step": 6100}, {"loss": 1.1106, "grad_norm": 0.7997331619262695, "learning_rate": 0.0002, "epoch": 3.9934640522875817, "step": 6110}, {"loss": 1.1244, "grad_norm": 0.7163209319114685, "learning_rate": 0.0002, "epoch": 4.0, "step": 6120}, {"eval_loss": 1.4113320112228394, "eval_runtime": 33.7199, "eval_samples_per_second": 12.93, "eval_steps_per_second": 1.631, "epoch": 4.0, "step": 6120}, {"loss": 1.0423, "grad_norm": 0.9527022838592529, "learning_rate": 0.0002, "epoch": 4.006535947712418, "step": 6130}, {"loss": 1.101, "grad_norm": 0.7603210210800171, "learning_rate": 0.0002, "epoch": 4.0130718954248366, "step": 6140}, {"loss": 1.1834, "grad_norm": 1.127387523651123, "learning_rate": 0.0002, "epoch": 4.019607843137255, "step": 6150}, {"loss": 1.0734, "grad_norm": 0.8290133476257324, "learning_rate": 0.0002, "epoch": 4.026143790849673, "step": 6160}, {"loss": 1.0785, "grad_norm": 0.9912241101264954, "learning_rate": 0.0002, "epoch": 4.032679738562091, "step": 6170}, {"loss": 1.0719, "grad_norm": 0.947005033493042, "learning_rate": 0.0002, "epoch": 4.03921568627451, "step": 6180}, {"loss": 1.0835, "grad_norm": 0.707466185092926, "learning_rate": 0.0002, "epoch": 4.045751633986928, "step": 6190}, {"loss": 1.1079, "grad_norm": 1.0604327917099, "learning_rate": 0.0002, "epoch": 4.052287581699346, "step": 6200}, {"loss": 1.0375, "grad_norm": 0.7848685383796692, "learning_rate": 0.0002, "epoch": 4.0588235294117645, "step": 6210}, {"loss": 1.1167, "grad_norm": 0.8475256562232971, "learning_rate": 0.0002, "epoch": 4.065359477124183, "step": 6220}, {"loss": 1.1104, "grad_norm": 0.9759448766708374, "learning_rate": 0.0002, "epoch": 4.071895424836601, "step": 6230}, {"loss": 1.1538, "grad_norm": 0.9324519038200378, "learning_rate": 0.0002, "epoch": 4.078431372549019, "step": 6240}, {"loss": 1.0817, "grad_norm": 0.8723901510238647, "learning_rate": 0.0002, "epoch": 4.084967320261438, "step": 6250}, {"loss": 1.0977, "grad_norm": 0.8343415856361389, "learning_rate": 0.0002, "epoch": 4.091503267973856, "step": 6260}, {"loss": 0.9887, "grad_norm": 0.7490310072898865, "learning_rate": 0.0002, "epoch": 4.098039215686274, "step": 6270}, {"loss": 1.2084, "grad_norm": 0.8961182832717896, "learning_rate": 0.0002, "epoch": 4.104575163398692, "step": 6280}, {"loss": 1.1349, "grad_norm": 0.7124854922294617, "learning_rate": 0.0002, "epoch": 4.111111111111111, "step": 6290}, {"loss": 1.0081, "grad_norm": 0.8338138461112976, "learning_rate": 0.0002, "epoch": 4.117647058823529, "step": 6300}, {"loss": 1.1091, "grad_norm": 0.8075833320617676, "learning_rate": 0.0002, "epoch": 4.124183006535947, "step": 6310}, {"loss": 1.0193, "grad_norm": 0.8069391846656799, "learning_rate": 0.0002, "epoch": 4.130718954248366, "step": 6320}, {"loss": 0.948, "grad_norm": 0.9567893147468567, "learning_rate": 0.0002, "epoch": 4.137254901960785, "step": 6330}, {"loss": 1.0241, "grad_norm": 1.2184662818908691, "learning_rate": 0.0002, "epoch": 4.143790849673203, "step": 6340}, {"loss": 1.0756, "grad_norm": 1.030976414680481, "learning_rate": 0.0002, "epoch": 4.150326797385621, "step": 6350}, {"loss": 1.1124, "grad_norm": 0.9749957323074341, "learning_rate": 0.0002, "epoch": 4.1568627450980395, "step": 6360}, {"loss": 1.1038, "grad_norm": 0.7089483141899109, "learning_rate": 0.0002, "epoch": 4.163398692810458, "step": 6370}, {"loss": 1.2175, "grad_norm": 1.1084946393966675, "learning_rate": 0.0002, "epoch": 4.169934640522876, "step": 6380}, {"loss": 1.0274, "grad_norm": 0.7998497486114502, "learning_rate": 0.0002, "epoch": 4.176470588235294, "step": 6390}, {"loss": 1.005, "grad_norm": 0.8997811675071716, "learning_rate": 0.0002, "epoch": 4.183006535947713, "step": 6400}, {"loss": 1.0704, "grad_norm": 0.8359479904174805, "learning_rate": 0.0002, "epoch": 4.189542483660131, "step": 6410}, {"loss": 1.1056, "grad_norm": 0.9087472558021545, "learning_rate": 0.0002, "epoch": 4.196078431372549, "step": 6420}, {"loss": 1.0657, "grad_norm": 1.1100451946258545, "learning_rate": 0.0002, "epoch": 4.2026143790849675, "step": 6430}, {"loss": 1.1443, "grad_norm": 0.9376999735832214, "learning_rate": 0.0002, "epoch": 4.209150326797386, "step": 6440}, {"loss": 1.0862, "grad_norm": 0.8179266452789307, "learning_rate": 0.0002, "epoch": 4.215686274509804, "step": 6450}, {"loss": 1.0679, "grad_norm": 0.9953271746635437, "learning_rate": 0.0002, "epoch": 4.222222222222222, "step": 6460}, {"loss": 1.1034, "grad_norm": 0.8476650714874268, "learning_rate": 0.0002, "epoch": 4.228758169934641, "step": 6470}, {"loss": 1.2512, "grad_norm": 0.8406323194503784, "learning_rate": 0.0002, "epoch": 4.235294117647059, "step": 6480}, {"loss": 1.057, "grad_norm": 0.819134533405304, "learning_rate": 0.0002, "epoch": 4.241830065359477, "step": 6490}, {"loss": 1.1082, "grad_norm": 0.7764983773231506, "learning_rate": 0.0002, "epoch": 4.248366013071895, "step": 6500}, {"loss": 1.1593, "grad_norm": 0.8252112865447998, "learning_rate": 0.0002, "epoch": 4.254901960784314, "step": 6510}, {"loss": 1.1369, "grad_norm": 0.7941019535064697, "learning_rate": 0.0002, "epoch": 4.261437908496732, "step": 6520}, {"loss": 1.0296, "grad_norm": 0.7673905491828918, "learning_rate": 0.0002, "epoch": 4.26797385620915, "step": 6530}, {"loss": 1.1387, "grad_norm": 0.8749890327453613, "learning_rate": 0.0002, "epoch": 4.2745098039215685, "step": 6540}, {"loss": 1.0595, "grad_norm": 0.7343207597732544, "learning_rate": 0.0002, "epoch": 4.281045751633987, "step": 6550}, {"loss": 1.1715, "grad_norm": 1.2786651849746704, "learning_rate": 0.0002, "epoch": 4.287581699346405, "step": 6560}, {"loss": 1.0514, "grad_norm": 1.316875696182251, "learning_rate": 0.0002, "epoch": 4.294117647058823, "step": 6570}, {"loss": 1.1125, "grad_norm": 0.8349189162254333, "learning_rate": 0.0002, "epoch": 4.300653594771242, "step": 6580}, {"loss": 1.0732, "grad_norm": 0.7510647177696228, "learning_rate": 0.0002, "epoch": 4.30718954248366, "step": 6590}, {"loss": 1.1387, "grad_norm": 0.932420551776886, "learning_rate": 0.0002, "epoch": 4.313725490196078, "step": 6600}, {"loss": 1.1115, "grad_norm": 0.8510616421699524, "learning_rate": 0.0002, "epoch": 4.3202614379084965, "step": 6610}, {"loss": 1.0957, "grad_norm": 0.7661547064781189, "learning_rate": 0.0002, "epoch": 4.326797385620915, "step": 6620}, {"loss": 1.2064, "grad_norm": 1.0370930433273315, "learning_rate": 0.0002, "epoch": 4.333333333333333, "step": 6630}, {"loss": 1.1064, "grad_norm": 0.9302158951759338, "learning_rate": 0.0002, "epoch": 4.339869281045751, "step": 6640}, {"loss": 0.968, "grad_norm": 0.9203811883926392, "learning_rate": 0.0002, "epoch": 4.34640522875817, "step": 6650}, {"loss": 1.0123, "grad_norm": 0.9986332654953003, "learning_rate": 0.0002, "epoch": 4.352941176470588, "step": 6660}, {"loss": 1.1079, "grad_norm": 0.8001713156700134, "learning_rate": 0.0002, "epoch": 4.359477124183006, "step": 6670}, {"loss": 1.0248, "grad_norm": 0.829714298248291, "learning_rate": 0.0002, "epoch": 4.366013071895424, "step": 6680}, {"loss": 1.0389, "grad_norm": 0.8253079056739807, "learning_rate": 0.0002, "epoch": 4.372549019607844, "step": 6690}, {"loss": 1.1087, "grad_norm": 0.824666440486908, "learning_rate": 0.0002, "epoch": 4.379084967320262, "step": 6700}, {"loss": 1.1968, "grad_norm": 0.8872972130775452, "learning_rate": 0.0002, "epoch": 4.38562091503268, "step": 6710}, {"loss": 1.0474, "grad_norm": 0.8729761838912964, "learning_rate": 0.0002, "epoch": 4.392156862745098, "step": 6720}, {"loss": 1.0961, "grad_norm": 1.1367264986038208, "learning_rate": 0.0002, "epoch": 4.398692810457517, "step": 6730}, {"loss": 1.0184, "grad_norm": 0.9699058532714844, "learning_rate": 0.0002, "epoch": 4.405228758169935, "step": 6740}, {"loss": 1.006, "grad_norm": 0.8266763687133789, "learning_rate": 0.0002, "epoch": 4.411764705882353, "step": 6750}, {"loss": 1.0735, "grad_norm": 1.0249767303466797, "learning_rate": 0.0002, "epoch": 4.4183006535947715, "step": 6760}, {"loss": 1.1726, "grad_norm": 0.73606938123703, "learning_rate": 0.0002, "epoch": 4.42483660130719, "step": 6770}, {"loss": 1.1037, "grad_norm": 1.4050679206848145, "learning_rate": 0.0002, "epoch": 4.431372549019608, "step": 6780}, {"loss": 1.1418, "grad_norm": 1.1114081144332886, "learning_rate": 0.0002, "epoch": 4.437908496732026, "step": 6790}, {"loss": 0.9682, "grad_norm": 0.8031067848205566, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6800}, {"loss": 1.0753, "grad_norm": 0.8513566851615906, "learning_rate": 0.0002, "epoch": 4.450980392156863, "step": 6810}, {"loss": 1.1852, "grad_norm": 1.332741379737854, "learning_rate": 0.0002, "epoch": 4.457516339869281, "step": 6820}, {"loss": 1.0966, "grad_norm": 1.5032578706741333, "learning_rate": 0.0002, "epoch": 4.4640522875816995, "step": 6830}, {"loss": 1.1124, "grad_norm": 0.7677283883094788, "learning_rate": 0.0002, "epoch": 4.470588235294118, "step": 6840}, {"loss": 1.1501, "grad_norm": 0.989148736000061, "learning_rate": 0.0002, "epoch": 4.477124183006536, "step": 6850}, {"loss": 1.2239, "grad_norm": 1.5316275358200073, "learning_rate": 0.0002, "epoch": 4.483660130718954, "step": 6860}, {"loss": 1.1171, "grad_norm": 0.9427124261856079, "learning_rate": 0.0002, "epoch": 4.490196078431373, "step": 6870}, {"loss": 1.1314, "grad_norm": 1.215287685394287, "learning_rate": 0.0002, "epoch": 4.496732026143791, "step": 6880}, {"loss": 1.0809, "grad_norm": 0.7286760210990906, "learning_rate": 0.0002, "epoch": 4.503267973856209, "step": 6890}, {"loss": 1.0179, "grad_norm": 0.874829888343811, "learning_rate": 0.0002, "epoch": 4.509803921568627, "step": 6900}, {"loss": 1.0233, "grad_norm": 0.8058359622955322, "learning_rate": 0.0002, "epoch": 4.516339869281046, "step": 6910}, {"loss": 1.0463, "grad_norm": 1.248195767402649, "learning_rate": 0.0002, "epoch": 4.522875816993464, "step": 6920}, {"loss": 1.0347, "grad_norm": 0.8033645749092102, "learning_rate": 0.0002, "epoch": 4.529411764705882, "step": 6930}, {"loss": 1.1068, "grad_norm": 1.7361950874328613, "learning_rate": 0.0002, "epoch": 4.5359477124183005, "step": 6940}, {"loss": 0.9856, "grad_norm": 0.8058095574378967, "learning_rate": 0.0002, "epoch": 4.542483660130719, "step": 6950}, {"loss": 1.0057, "grad_norm": 1.254089593887329, "learning_rate": 0.0002, "epoch": 4.549019607843137, "step": 6960}, {"loss": 1.1723, "grad_norm": 0.9180455803871155, "learning_rate": 0.0002, "epoch": 4.555555555555555, "step": 6970}, {"loss": 1.0559, "grad_norm": 0.6677682399749756, "learning_rate": 0.0002, "epoch": 4.562091503267974, "step": 6980}, {"loss": 1.0453, "grad_norm": 0.8127354383468628, "learning_rate": 0.0002, "epoch": 4.568627450980392, "step": 6990}, {"loss": 1.0828, "grad_norm": 1.0263001918792725, "learning_rate": 0.0002, "epoch": 4.57516339869281, "step": 7000}, {"loss": 1.0703, "grad_norm": 0.9641909003257751, "learning_rate": 0.0002, "epoch": 4.5816993464052285, "step": 7010}, {"loss": 1.179, "grad_norm": 0.9440861344337463, "learning_rate": 0.0002, "epoch": 4.588235294117647, "step": 7020}, {"loss": 1.0931, "grad_norm": 0.9539011716842651, "learning_rate": 0.0002, "epoch": 4.594771241830065, "step": 7030}, {"loss": 1.0963, "grad_norm": 1.0449910163879395, "learning_rate": 0.0002, "epoch": 4.601307189542483, "step": 7040}, {"loss": 0.9944, "grad_norm": 0.8766893744468689, "learning_rate": 0.0002, "epoch": 4.607843137254902, "step": 7050}, {"loss": 1.0169, "grad_norm": 0.6983462572097778, "learning_rate": 0.0002, "epoch": 4.61437908496732, "step": 7060}, {"loss": 1.1778, "grad_norm": 0.9505505561828613, "learning_rate": 0.0002, "epoch": 4.620915032679738, "step": 7070}, {"loss": 1.121, "grad_norm": 1.2506657838821411, "learning_rate": 0.0002, "epoch": 4.627450980392156, "step": 7080}, {"loss": 1.1329, "grad_norm": 0.9602801203727722, "learning_rate": 0.0002, "epoch": 4.633986928104575, "step": 7090}, {"loss": 1.1499, "grad_norm": 0.7398977875709534, "learning_rate": 0.0002, "epoch": 4.640522875816993, "step": 7100}, {"loss": 1.0769, "grad_norm": 1.3862425088882446, "learning_rate": 0.0002, "epoch": 4.647058823529412, "step": 7110}, {"loss": 1.0571, "grad_norm": 1.1451990604400635, "learning_rate": 0.0002, "epoch": 4.65359477124183, "step": 7120}, {"loss": 1.1271, "grad_norm": 0.9010422229766846, "learning_rate": 0.0002, "epoch": 4.660130718954249, "step": 7130}, {"loss": 1.0165, "grad_norm": 0.7102518081665039, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 7140}, {"loss": 1.0819, "grad_norm": 0.7963796257972717, "learning_rate": 0.0002, "epoch": 4.673202614379085, "step": 7150}, {"loss": 1.1114, "grad_norm": 0.7726007699966431, "learning_rate": 0.0002, "epoch": 4.6797385620915035, "step": 7160}, {"loss": 1.2088, "grad_norm": 0.8097564578056335, "learning_rate": 0.0002, "epoch": 4.686274509803922, "step": 7170}, {"loss": 1.1386, "grad_norm": 0.9070925116539001, "learning_rate": 0.0002, "epoch": 4.69281045751634, "step": 7180}, {"loss": 1.0315, "grad_norm": 0.7543528079986572, "learning_rate": 0.0002, "epoch": 4.699346405228758, "step": 7190}, {"loss": 1.0984, "grad_norm": 0.9900904893875122, "learning_rate": 0.0002, "epoch": 4.705882352941177, "step": 7200}, {"loss": 1.1552, "grad_norm": 0.8033412098884583, "learning_rate": 0.0002, "epoch": 4.712418300653595, "step": 7210}, {"loss": 1.1773, "grad_norm": 0.8440839052200317, "learning_rate": 0.0002, "epoch": 4.718954248366013, "step": 7220}, {"loss": 1.1258, "grad_norm": 0.9325555562973022, "learning_rate": 0.0002, "epoch": 4.7254901960784315, "step": 7230}, {"loss": 1.1384, "grad_norm": 0.7881146669387817, "learning_rate": 0.0002, "epoch": 4.73202614379085, "step": 7240}, {"loss": 1.1219, "grad_norm": 0.884453296661377, "learning_rate": 0.0002, "epoch": 4.738562091503268, "step": 7250}, {"loss": 1.1036, "grad_norm": 0.9274539351463318, "learning_rate": 0.0002, "epoch": 4.745098039215686, "step": 7260}, {"loss": 1.0906, "grad_norm": 1.2367479801177979, "learning_rate": 0.0002, "epoch": 4.751633986928105, "step": 7270}, {"loss": 1.0741, "grad_norm": 0.9499821066856384, "learning_rate": 0.0002, "epoch": 4.758169934640523, "step": 7280}, {"loss": 1.1625, "grad_norm": 2.1918580532073975, "learning_rate": 0.0002, "epoch": 4.764705882352941, "step": 7290}, {"loss": 0.954, "grad_norm": 0.8221880793571472, "learning_rate": 0.0002, "epoch": 4.771241830065359, "step": 7300}, {"loss": 1.1358, "grad_norm": 0.871972918510437, "learning_rate": 0.0002, "epoch": 4.777777777777778, "step": 7310}, {"loss": 1.0599, "grad_norm": 0.8034510612487793, "learning_rate": 0.0002, "epoch": 4.784313725490196, "step": 7320}, {"loss": 1.1059, "grad_norm": 0.8959605693817139, "learning_rate": 0.0002, "epoch": 4.790849673202614, "step": 7330}, {"loss": 1.0176, "grad_norm": 1.2326215505599976, "learning_rate": 0.0002, "epoch": 4.7973856209150325, "step": 7340}, {"loss": 1.1095, "grad_norm": 0.9725791811943054, "learning_rate": 0.0002, "epoch": 4.803921568627451, "step": 7350}, {"loss": 1.1229, "grad_norm": 0.7240816354751587, "learning_rate": 0.0002, "epoch": 4.810457516339869, "step": 7360}, {"loss": 1.0669, "grad_norm": 0.8265769481658936, "learning_rate": 0.0002, "epoch": 4.816993464052287, "step": 7370}, {"loss": 1.042, "grad_norm": 0.8888696432113647, "learning_rate": 0.0002, "epoch": 4.823529411764706, "step": 7380}, {"loss": 1.0981, "grad_norm": 0.7776556015014648, "learning_rate": 0.0002, "epoch": 4.830065359477124, "step": 7390}, {"loss": 1.0819, "grad_norm": 0.8772371411323547, "learning_rate": 0.0002, "epoch": 4.836601307189542, "step": 7400}, {"loss": 1.0819, "grad_norm": 0.9786531925201416, "learning_rate": 0.0002, "epoch": 4.8431372549019605, "step": 7410}, {"loss": 1.1358, "grad_norm": 0.9059745073318481, "learning_rate": 0.0002, "epoch": 4.849673202614379, "step": 7420}, {"loss": 1.0324, "grad_norm": 0.7422552108764648, "learning_rate": 0.0002, "epoch": 4.856209150326797, "step": 7430}, {"loss": 1.0423, "grad_norm": 1.3040380477905273, "learning_rate": 0.0002, "epoch": 4.862745098039216, "step": 7440}, {"loss": 1.1161, "grad_norm": 1.3278473615646362, "learning_rate": 0.0002, "epoch": 4.8692810457516345, "step": 7450}, {"loss": 1.0713, "grad_norm": 1.2705849409103394, "learning_rate": 0.0002, "epoch": 4.875816993464053, "step": 7460}, {"loss": 1.0034, "grad_norm": 0.8837892413139343, "learning_rate": 0.0002, "epoch": 4.882352941176471, "step": 7470}, {"loss": 1.1716, "grad_norm": 0.8670691251754761, "learning_rate": 0.0002, "epoch": 4.888888888888889, "step": 7480}, {"loss": 1.1723, "grad_norm": 0.9662758111953735, "learning_rate": 0.0002, "epoch": 4.895424836601308, "step": 7490}, {"loss": 1.1056, "grad_norm": 0.8188302516937256, "learning_rate": 0.0002, "epoch": 4.901960784313726, "step": 7500}, {"loss": 1.0419, "grad_norm": 0.769442617893219, "learning_rate": 0.0002, "epoch": 4.908496732026144, "step": 7510}, {"loss": 1.1671, "grad_norm": 1.1465084552764893, "learning_rate": 0.0002, "epoch": 4.915032679738562, "step": 7520}, {"loss": 1.0768, "grad_norm": 1.253214955329895, "learning_rate": 0.0002, "epoch": 4.921568627450981, "step": 7530}, {"loss": 1.011, "grad_norm": 0.7922375202178955, "learning_rate": 0.0002, "epoch": 4.928104575163399, "step": 7540}, {"loss": 1.1256, "grad_norm": 0.8306851387023926, "learning_rate": 0.0002, "epoch": 4.934640522875817, "step": 7550}, {"loss": 1.206, "grad_norm": 0.8486151099205017, "learning_rate": 0.0002, "epoch": 4.9411764705882355, "step": 7560}, {"loss": 1.0161, "grad_norm": 1.2601467370986938, "learning_rate": 0.0002, "epoch": 4.947712418300654, "step": 7570}, {"loss": 1.1078, "grad_norm": 0.7980747818946838, "learning_rate": 0.0002, "epoch": 4.954248366013072, "step": 7580}, {"loss": 1.0607, "grad_norm": 0.8653254508972168, "learning_rate": 0.0002, "epoch": 4.96078431372549, "step": 7590}, {"loss": 1.0292, "grad_norm": 0.9680571556091309, "learning_rate": 0.0002, "epoch": 4.967320261437909, "step": 7600}, {"loss": 1.1795, "grad_norm": 0.9554466605186462, "learning_rate": 0.0002, "epoch": 4.973856209150327, "step": 7610}, {"loss": 1.0935, "grad_norm": 1.3693897724151611, "learning_rate": 0.0002, "epoch": 4.980392156862745, "step": 7620}, {"loss": 1.0838, "grad_norm": 0.7809282541275024, "learning_rate": 0.0002, "epoch": 4.9869281045751634, "step": 7630}, {"loss": 1.0844, "grad_norm": 0.7528006434440613, "learning_rate": 0.0002, "epoch": 4.993464052287582, "step": 7640}, {"loss": 0.9951, "grad_norm": 1.7491309642791748, "learning_rate": 0.0002, "epoch": 5.0, "step": 7650}]}
+{"epoch": 6.0, "step": 9180, "epoch_duration": 1806.2337534427643, "total_accumulated_duration": 10291.616841077805, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}, {"eval_loss": 1.4131312370300293, "eval_runtime": 31.8967, "eval_samples_per_second": 13.669, "eval_steps_per_second": 1.724, "epoch": 3.0, "step": 4590}, {"loss": 1.1283, "grad_norm": 0.6926420331001282, "learning_rate": 0.0002, "epoch": 3.0065359477124183, "step": 4600}, {"loss": 1.1537, "grad_norm": 0.6656355857849121, "learning_rate": 0.0002, "epoch": 3.0130718954248366, "step": 4610}, {"loss": 1.308, "grad_norm": 0.9901936650276184, "learning_rate": 0.0002, "epoch": 3.019607843137255, "step": 4620}, {"loss": 1.22, "grad_norm": 0.6713474988937378, "learning_rate": 0.0002, "epoch": 3.026143790849673, "step": 4630}, {"loss": 1.2249, "grad_norm": 0.6199324131011963, "learning_rate": 0.0002, "epoch": 3.0326797385620914, "step": 4640}, {"loss": 1.242, "grad_norm": 0.7180785536766052, "learning_rate": 0.0002, "epoch": 3.0392156862745097, "step": 4650}, {"loss": 1.1349, "grad_norm": 0.8256588578224182, "learning_rate": 0.0002, "epoch": 3.045751633986928, "step": 4660}, {"loss": 1.1431, "grad_norm": 0.6637389063835144, "learning_rate": 0.0002, "epoch": 3.052287581699346, "step": 4670}, {"loss": 1.1096, "grad_norm": 0.6980698108673096, "learning_rate": 0.0002, "epoch": 3.0588235294117645, "step": 4680}, {"loss": 1.196, "grad_norm": 0.8091534972190857, "learning_rate": 0.0002, "epoch": 3.065359477124183, "step": 4690}, {"loss": 1.1652, "grad_norm": 0.5715174078941345, "learning_rate": 0.0002, "epoch": 3.0718954248366015, "step": 4700}, {"loss": 1.1427, "grad_norm": 0.735639750957489, "learning_rate": 0.0002, "epoch": 3.0784313725490198, "step": 4710}, {"loss": 1.1522, "grad_norm": 0.7619708180427551, "learning_rate": 0.0002, "epoch": 3.084967320261438, "step": 4720}, {"loss": 1.0853, "grad_norm": 1.263566017150879, "learning_rate": 0.0002, "epoch": 3.0915032679738563, "step": 4730}, {"loss": 1.1348, "grad_norm": 0.6600871682167053, "learning_rate": 0.0002, "epoch": 3.0980392156862746, "step": 4740}, {"loss": 1.1766, "grad_norm": 0.717792809009552, "learning_rate": 0.0002, "epoch": 3.104575163398693, "step": 4750}, {"loss": 1.088, "grad_norm": 0.853714644908905, "learning_rate": 0.0002, "epoch": 3.111111111111111, "step": 4760}, {"loss": 1.2031, "grad_norm": 1.1004153490066528, "learning_rate": 0.0002, "epoch": 3.1176470588235294, "step": 4770}, {"loss": 1.3295, "grad_norm": 0.8566235899925232, "learning_rate": 0.0002, "epoch": 3.1241830065359477, "step": 4780}, {"loss": 1.2436, "grad_norm": 0.8315296173095703, "learning_rate": 0.0002, "epoch": 3.130718954248366, "step": 4790}, {"loss": 1.32, "grad_norm": 0.8020524978637695, "learning_rate": 0.0002, "epoch": 3.1372549019607843, "step": 4800}, {"loss": 1.1238, "grad_norm": 0.7564275860786438, "learning_rate": 0.0002, "epoch": 3.1437908496732025, "step": 4810}, {"loss": 1.1244, "grad_norm": 0.9077776670455933, "learning_rate": 0.0002, "epoch": 3.150326797385621, "step": 4820}, {"loss": 1.1399, "grad_norm": 0.6323099732398987, "learning_rate": 0.0002, "epoch": 3.156862745098039, "step": 4830}, {"loss": 1.1983, "grad_norm": 0.6625368595123291, "learning_rate": 0.0002, "epoch": 3.1633986928104574, "step": 4840}, {"loss": 1.066, "grad_norm": 0.8119261860847473, "learning_rate": 0.0002, "epoch": 3.1699346405228757, "step": 4850}, {"loss": 1.0224, "grad_norm": 0.6399450898170471, "learning_rate": 0.0002, "epoch": 3.176470588235294, "step": 4860}, {"loss": 1.2181, "grad_norm": 1.0659016370773315, "learning_rate": 0.0002, "epoch": 3.183006535947712, "step": 4870}, {"loss": 1.2914, "grad_norm": 0.8040369749069214, "learning_rate": 0.0002, "epoch": 3.189542483660131, "step": 4880}, {"loss": 1.1996, "grad_norm": 0.7784733176231384, "learning_rate": 0.0002, "epoch": 3.196078431372549, "step": 4890}, {"loss": 1.2051, "grad_norm": 0.9660294651985168, "learning_rate": 0.0002, "epoch": 3.2026143790849675, "step": 4900}, {"loss": 1.0419, "grad_norm": 1.0676977634429932, "learning_rate": 0.0002, "epoch": 3.2091503267973858, "step": 4910}, {"loss": 1.0083, "grad_norm": 0.5877565741539001, "learning_rate": 0.0002, "epoch": 3.215686274509804, "step": 4920}, {"loss": 1.1046, "grad_norm": 0.6164032816886902, "learning_rate": 0.0002, "epoch": 3.2222222222222223, "step": 4930}, {"loss": 1.1079, "grad_norm": 0.7627606987953186, "learning_rate": 0.0002, "epoch": 3.2287581699346406, "step": 4940}, {"loss": 1.2453, "grad_norm": 0.7442803978919983, "learning_rate": 0.0002, "epoch": 3.235294117647059, "step": 4950}, {"loss": 1.1087, "grad_norm": 0.7277812361717224, "learning_rate": 0.0002, "epoch": 3.241830065359477, "step": 4960}, {"loss": 1.2237, "grad_norm": 1.0301902294158936, "learning_rate": 0.0002, "epoch": 3.2483660130718954, "step": 4970}, {"loss": 1.1466, "grad_norm": 0.7798232436180115, "learning_rate": 0.0002, "epoch": 3.2549019607843137, "step": 4980}, {"loss": 1.2142, "grad_norm": 1.210265874862671, "learning_rate": 0.0002, "epoch": 3.261437908496732, "step": 4990}, {"loss": 1.1557, "grad_norm": 0.6677713990211487, "learning_rate": 0.0002, "epoch": 3.2679738562091503, "step": 5000}, {"loss": 1.3294, "grad_norm": 1.0524500608444214, "learning_rate": 0.0002, "epoch": 3.2745098039215685, "step": 5010}, {"loss": 1.1939, "grad_norm": 0.7091745734214783, "learning_rate": 0.0002, "epoch": 3.281045751633987, "step": 5020}, {"loss": 1.1891, "grad_norm": 0.8523224592208862, "learning_rate": 0.0002, "epoch": 3.287581699346405, "step": 5030}, {"loss": 1.1925, "grad_norm": 0.6120608448982239, "learning_rate": 0.0002, "epoch": 3.2941176470588234, "step": 5040}, {"loss": 1.0603, "grad_norm": 0.7437472939491272, "learning_rate": 0.0002, "epoch": 3.3006535947712417, "step": 5050}, {"loss": 1.1295, "grad_norm": 0.7611715197563171, "learning_rate": 0.0002, "epoch": 3.30718954248366, "step": 5060}, {"loss": 1.0531, "grad_norm": 0.7249704003334045, "learning_rate": 0.0002, "epoch": 3.313725490196078, "step": 5070}, {"loss": 1.2292, "grad_norm": 0.7316247820854187, "learning_rate": 0.0002, "epoch": 3.3202614379084965, "step": 5080}, {"loss": 1.1974, "grad_norm": 0.562412440776825, "learning_rate": 0.0002, "epoch": 3.326797385620915, "step": 5090}, {"loss": 1.0736, "grad_norm": 0.7052176594734192, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 5100}, {"loss": 1.122, "grad_norm": 0.7714211344718933, "learning_rate": 0.0002, "epoch": 3.3398692810457518, "step": 5110}, {"loss": 1.1684, "grad_norm": 1.0436055660247803, "learning_rate": 0.0002, "epoch": 3.34640522875817, "step": 5120}, {"loss": 1.0945, "grad_norm": 0.8867271542549133, "learning_rate": 0.0002, "epoch": 3.3529411764705883, "step": 5130}, {"loss": 1.159, "grad_norm": 0.8371267914772034, "learning_rate": 0.0002, "epoch": 3.3594771241830066, "step": 5140}, {"loss": 1.1073, "grad_norm": 0.7257837057113647, "learning_rate": 0.0002, "epoch": 3.366013071895425, "step": 5150}, {"loss": 1.1162, "grad_norm": 0.7102002501487732, "learning_rate": 0.0002, "epoch": 3.372549019607843, "step": 5160}, {"loss": 1.2056, "grad_norm": 0.7636350393295288, "learning_rate": 0.0002, "epoch": 3.3790849673202614, "step": 5170}, {"loss": 1.0708, "grad_norm": 0.6887359619140625, "learning_rate": 0.0002, "epoch": 3.3856209150326797, "step": 5180}, {"loss": 1.3807, "grad_norm": 0.8141424655914307, "learning_rate": 0.0002, "epoch": 3.392156862745098, "step": 5190}, {"loss": 1.1986, "grad_norm": 0.694423496723175, "learning_rate": 0.0002, "epoch": 3.3986928104575163, "step": 5200}, {"loss": 1.2945, "grad_norm": 0.914013683795929, "learning_rate": 0.0002, "epoch": 3.4052287581699345, "step": 5210}, {"loss": 1.1413, "grad_norm": 0.8503239750862122, "learning_rate": 0.0002, "epoch": 3.411764705882353, "step": 5220}, {"loss": 1.2696, "grad_norm": 0.6196836233139038, "learning_rate": 0.0002, "epoch": 3.418300653594771, "step": 5230}, {"loss": 1.2431, "grad_norm": 1.0760811567306519, "learning_rate": 0.0002, "epoch": 3.4248366013071894, "step": 5240}, {"loss": 1.1686, "grad_norm": 0.6524698138237, "learning_rate": 0.0002, "epoch": 3.431372549019608, "step": 5250}, {"loss": 1.2012, "grad_norm": 0.674467921257019, "learning_rate": 0.0002, "epoch": 3.4379084967320264, "step": 5260}, {"loss": 1.1015, "grad_norm": 0.7690372467041016, "learning_rate": 0.0002, "epoch": 3.4444444444444446, "step": 5270}, {"loss": 1.2511, "grad_norm": 0.8751813769340515, "learning_rate": 0.0002, "epoch": 3.450980392156863, "step": 5280}, {"loss": 1.1841, "grad_norm": 0.750407874584198, "learning_rate": 0.0002, "epoch": 3.457516339869281, "step": 5290}, {"loss": 1.0605, "grad_norm": 0.5991823077201843, "learning_rate": 0.0002, "epoch": 3.4640522875816995, "step": 5300}, {"loss": 1.2347, "grad_norm": 1.0164772272109985, "learning_rate": 0.0002, "epoch": 3.4705882352941178, "step": 5310}, {"loss": 1.2354, "grad_norm": 0.8704105019569397, "learning_rate": 0.0002, "epoch": 3.477124183006536, "step": 5320}, {"loss": 1.2169, "grad_norm": 0.709102213382721, "learning_rate": 0.0002, "epoch": 3.4836601307189543, "step": 5330}, {"loss": 1.2425, "grad_norm": 0.6273632049560547, "learning_rate": 0.0002, "epoch": 3.4901960784313726, "step": 5340}, {"loss": 1.1585, "grad_norm": 0.6807359457015991, "learning_rate": 0.0002, "epoch": 3.496732026143791, "step": 5350}, {"loss": 1.131, "grad_norm": 0.7085188627243042, "learning_rate": 0.0002, "epoch": 3.503267973856209, "step": 5360}, {"loss": 1.1159, "grad_norm": 0.6938307881355286, "learning_rate": 0.0002, "epoch": 3.5098039215686274, "step": 5370}, {"loss": 1.1397, "grad_norm": 0.8544146418571472, "learning_rate": 0.0002, "epoch": 3.5163398692810457, "step": 5380}, {"loss": 1.2181, "grad_norm": 0.7889642119407654, "learning_rate": 0.0002, "epoch": 3.522875816993464, "step": 5390}, {"loss": 1.1691, "grad_norm": 0.7858421206474304, "learning_rate": 0.0002, "epoch": 3.5294117647058822, "step": 5400}, {"loss": 1.2374, "grad_norm": 0.8547123074531555, "learning_rate": 0.0002, "epoch": 3.5359477124183005, "step": 5410}, {"loss": 1.196, "grad_norm": 0.8218181133270264, "learning_rate": 0.0002, "epoch": 3.542483660130719, "step": 5420}, {"loss": 1.1961, "grad_norm": 1.153623342514038, "learning_rate": 0.0002, "epoch": 3.549019607843137, "step": 5430}, {"loss": 1.156, "grad_norm": 1.1321099996566772, "learning_rate": 0.0002, "epoch": 3.5555555555555554, "step": 5440}, {"loss": 1.2224, "grad_norm": 0.9495334029197693, "learning_rate": 0.0002, "epoch": 3.5620915032679736, "step": 5450}, {"loss": 1.2869, "grad_norm": 0.8743821978569031, "learning_rate": 0.0002, "epoch": 3.568627450980392, "step": 5460}, {"loss": 1.1018, "grad_norm": 0.7513086795806885, "learning_rate": 0.0002, "epoch": 3.57516339869281, "step": 5470}, {"loss": 1.1082, "grad_norm": 1.0139480829238892, "learning_rate": 0.0002, "epoch": 3.581699346405229, "step": 5480}, {"loss": 1.1706, "grad_norm": 0.6615135073661804, "learning_rate": 0.0002, "epoch": 3.588235294117647, "step": 5490}, {"loss": 1.3906, "grad_norm": 1.180798888206482, "learning_rate": 0.0002, "epoch": 3.5947712418300655, "step": 5500}, {"loss": 1.2391, "grad_norm": 0.7085279226303101, "learning_rate": 0.0002, "epoch": 3.6013071895424837, "step": 5510}, {"loss": 1.1623, "grad_norm": 0.540268063545227, "learning_rate": 0.0002, "epoch": 3.607843137254902, "step": 5520}, {"loss": 1.2132, "grad_norm": 0.7905671000480652, "learning_rate": 0.0002, "epoch": 3.6143790849673203, "step": 5530}, {"loss": 1.2731, "grad_norm": 0.8457717299461365, "learning_rate": 0.0002, "epoch": 3.6209150326797386, "step": 5540}, {"loss": 1.1799, "grad_norm": 0.7102677822113037, "learning_rate": 0.0002, "epoch": 3.627450980392157, "step": 5550}, {"loss": 1.2394, "grad_norm": 0.7179514765739441, "learning_rate": 0.0002, "epoch": 3.633986928104575, "step": 5560}, {"loss": 1.2019, "grad_norm": 1.0854148864746094, "learning_rate": 0.0002, "epoch": 3.6405228758169934, "step": 5570}, {"loss": 1.1986, "grad_norm": 0.8209951519966125, "learning_rate": 0.0002, "epoch": 3.6470588235294117, "step": 5580}, {"loss": 1.2289, "grad_norm": 0.6944138407707214, "learning_rate": 0.0002, "epoch": 3.65359477124183, "step": 5590}, {"loss": 1.3226, "grad_norm": 0.7675473093986511, "learning_rate": 0.0002, "epoch": 3.6601307189542482, "step": 5600}, {"loss": 1.2866, "grad_norm": 0.6683364510536194, "learning_rate": 0.0002, "epoch": 3.6666666666666665, "step": 5610}, {"loss": 1.1099, "grad_norm": 0.7920727133750916, "learning_rate": 0.0002, "epoch": 3.6732026143790852, "step": 5620}, {"loss": 1.2287, "grad_norm": 0.9440218806266785, "learning_rate": 0.0002, "epoch": 3.6797385620915035, "step": 5630}, {"loss": 1.2444, "grad_norm": 0.6600824594497681, "learning_rate": 0.0002, "epoch": 3.686274509803922, "step": 5640}, {"loss": 1.191, "grad_norm": 0.6860619187355042, "learning_rate": 0.0002, "epoch": 3.69281045751634, "step": 5650}, {"loss": 1.1914, "grad_norm": 0.6579713225364685, "learning_rate": 0.0002, "epoch": 3.6993464052287583, "step": 5660}, {"loss": 1.1464, "grad_norm": 0.661081075668335, "learning_rate": 0.0002, "epoch": 3.7058823529411766, "step": 5670}, {"loss": 1.289, "grad_norm": 1.0968825817108154, "learning_rate": 0.0002, "epoch": 3.712418300653595, "step": 5680}, {"loss": 1.192, "grad_norm": 0.8066844940185547, "learning_rate": 0.0002, "epoch": 3.718954248366013, "step": 5690}, {"loss": 1.2322, "grad_norm": 0.8341682553291321, "learning_rate": 0.0002, "epoch": 3.7254901960784315, "step": 5700}, {"loss": 1.1473, "grad_norm": 0.6682852506637573, "learning_rate": 0.0002, "epoch": 3.7320261437908497, "step": 5710}, {"loss": 1.1566, "grad_norm": 0.898595929145813, "learning_rate": 0.0002, "epoch": 3.738562091503268, "step": 5720}, {"loss": 1.0919, "grad_norm": 0.6876054406166077, "learning_rate": 0.0002, "epoch": 3.7450980392156863, "step": 5730}, {"loss": 1.2302, "grad_norm": 0.7817103266716003, "learning_rate": 0.0002, "epoch": 3.7516339869281046, "step": 5740}, {"loss": 1.2439, "grad_norm": 0.5840168595314026, "learning_rate": 0.0002, "epoch": 3.758169934640523, "step": 5750}, {"loss": 1.1279, "grad_norm": 0.6263918876647949, "learning_rate": 0.0002, "epoch": 3.764705882352941, "step": 5760}, {"loss": 1.2023, "grad_norm": 0.7948952317237854, "learning_rate": 0.0002, "epoch": 3.7712418300653594, "step": 5770}, {"loss": 1.149, "grad_norm": 0.6700998544692993, "learning_rate": 0.0002, "epoch": 3.7777777777777777, "step": 5780}, {"loss": 1.3207, "grad_norm": 1.1169519424438477, "learning_rate": 0.0002, "epoch": 3.784313725490196, "step": 5790}, {"loss": 1.064, "grad_norm": 0.8354471325874329, "learning_rate": 0.0002, "epoch": 3.7908496732026142, "step": 5800}, {"loss": 1.2104, "grad_norm": 0.6304181814193726, "learning_rate": 0.0002, "epoch": 3.7973856209150325, "step": 5810}, {"loss": 1.2059, "grad_norm": 0.6919655799865723, "learning_rate": 0.0002, "epoch": 3.803921568627451, "step": 5820}, {"loss": 1.217, "grad_norm": 0.600385844707489, "learning_rate": 0.0002, "epoch": 3.810457516339869, "step": 5830}, {"loss": 1.2324, "grad_norm": 0.8406319618225098, "learning_rate": 0.0002, "epoch": 3.8169934640522873, "step": 5840}, {"loss": 1.2418, "grad_norm": 0.7594282031059265, "learning_rate": 0.0002, "epoch": 3.8235294117647056, "step": 5850}, {"loss": 1.1903, "grad_norm": 0.8179879784584045, "learning_rate": 0.0002, "epoch": 3.8300653594771243, "step": 5860}, {"loss": 1.255, "grad_norm": 1.141430377960205, "learning_rate": 0.0002, "epoch": 3.8366013071895426, "step": 5870}, {"loss": 1.1467, "grad_norm": 0.6595550775527954, "learning_rate": 0.0002, "epoch": 3.843137254901961, "step": 5880}, {"loss": 1.2378, "grad_norm": 0.7499435544013977, "learning_rate": 0.0002, "epoch": 3.849673202614379, "step": 5890}, {"loss": 1.217, "grad_norm": 0.7851517200469971, "learning_rate": 0.0002, "epoch": 3.8562091503267975, "step": 5900}, {"loss": 1.162, "grad_norm": 1.0533545017242432, "learning_rate": 0.0002, "epoch": 3.8627450980392157, "step": 5910}, {"loss": 1.3576, "grad_norm": 0.960086464881897, "learning_rate": 0.0002, "epoch": 3.869281045751634, "step": 5920}, {"loss": 1.151, "grad_norm": 0.9952049851417542, "learning_rate": 0.0002, "epoch": 3.8758169934640523, "step": 5930}, {"loss": 1.2027, "grad_norm": 0.7884191274642944, "learning_rate": 0.0002, "epoch": 3.8823529411764706, "step": 5940}, {"loss": 1.1796, "grad_norm": 0.7461766600608826, "learning_rate": 0.0002, "epoch": 3.888888888888889, "step": 5950}, {"loss": 1.2251, "grad_norm": 0.9594355821609497, "learning_rate": 0.0002, "epoch": 3.895424836601307, "step": 5960}, {"loss": 1.1164, "grad_norm": 0.8179471492767334, "learning_rate": 0.0002, "epoch": 3.9019607843137254, "step": 5970}, {"loss": 1.2421, "grad_norm": 0.8240267634391785, "learning_rate": 0.0002, "epoch": 3.9084967320261437, "step": 5980}, {"loss": 1.3076, "grad_norm": 0.7462618350982666, "learning_rate": 0.0002, "epoch": 3.915032679738562, "step": 5990}, {"loss": 1.2124, "grad_norm": 0.711207389831543, "learning_rate": 0.0002, "epoch": 3.9215686274509802, "step": 6000}, {"loss": 1.2119, "grad_norm": 0.6910956501960754, "learning_rate": 0.0002, "epoch": 3.928104575163399, "step": 6010}, {"loss": 1.2127, "grad_norm": 0.749093770980835, "learning_rate": 0.0002, "epoch": 3.9346405228758172, "step": 6020}, {"loss": 1.1542, "grad_norm": 1.3332762718200684, "learning_rate": 0.0002, "epoch": 3.9411764705882355, "step": 6030}, {"loss": 1.1442, "grad_norm": 0.71457439661026, "learning_rate": 0.0002, "epoch": 3.947712418300654, "step": 6040}, {"loss": 1.339, "grad_norm": 1.1205238103866577, "learning_rate": 0.0002, "epoch": 3.954248366013072, "step": 6050}, {"loss": 1.2962, "grad_norm": 0.6958928108215332, "learning_rate": 0.0002, "epoch": 3.9607843137254903, "step": 6060}, {"loss": 1.1802, "grad_norm": 0.7518056035041809, "learning_rate": 0.0002, "epoch": 3.9673202614379086, "step": 6070}, {"loss": 1.1179, "grad_norm": 0.8010755777359009, "learning_rate": 0.0002, "epoch": 3.973856209150327, "step": 6080}, {"loss": 1.2867, "grad_norm": 0.7492658495903015, "learning_rate": 0.0002, "epoch": 3.980392156862745, "step": 6090}, {"loss": 1.2113, "grad_norm": 0.900704562664032, "learning_rate": 0.0002, "epoch": 3.9869281045751634, "step": 6100}, {"loss": 1.1106, "grad_norm": 0.7997331619262695, "learning_rate": 0.0002, "epoch": 3.9934640522875817, "step": 6110}, {"loss": 1.1244, "grad_norm": 0.7163209319114685, "learning_rate": 0.0002, "epoch": 4.0, "step": 6120}, {"eval_loss": 1.4113320112228394, "eval_runtime": 33.7199, "eval_samples_per_second": 12.93, "eval_steps_per_second": 1.631, "epoch": 4.0, "step": 6120}, {"loss": 1.0423, "grad_norm": 0.9527022838592529, "learning_rate": 0.0002, "epoch": 4.006535947712418, "step": 6130}, {"loss": 1.101, "grad_norm": 0.7603210210800171, "learning_rate": 0.0002, "epoch": 4.0130718954248366, "step": 6140}, {"loss": 1.1834, "grad_norm": 1.127387523651123, "learning_rate": 0.0002, "epoch": 4.019607843137255, "step": 6150}, {"loss": 1.0734, "grad_norm": 0.8290133476257324, "learning_rate": 0.0002, "epoch": 4.026143790849673, "step": 6160}, {"loss": 1.0785, "grad_norm": 0.9912241101264954, "learning_rate": 0.0002, "epoch": 4.032679738562091, "step": 6170}, {"loss": 1.0719, "grad_norm": 0.947005033493042, "learning_rate": 0.0002, "epoch": 4.03921568627451, "step": 6180}, {"loss": 1.0835, "grad_norm": 0.707466185092926, "learning_rate": 0.0002, "epoch": 4.045751633986928, "step": 6190}, {"loss": 1.1079, "grad_norm": 1.0604327917099, "learning_rate": 0.0002, "epoch": 4.052287581699346, "step": 6200}, {"loss": 1.0375, "grad_norm": 0.7848685383796692, "learning_rate": 0.0002, "epoch": 4.0588235294117645, "step": 6210}, {"loss": 1.1167, "grad_norm": 0.8475256562232971, "learning_rate": 0.0002, "epoch": 4.065359477124183, "step": 6220}, {"loss": 1.1104, "grad_norm": 0.9759448766708374, "learning_rate": 0.0002, "epoch": 4.071895424836601, "step": 6230}, {"loss": 1.1538, "grad_norm": 0.9324519038200378, "learning_rate": 0.0002, "epoch": 4.078431372549019, "step": 6240}, {"loss": 1.0817, "grad_norm": 0.8723901510238647, "learning_rate": 0.0002, "epoch": 4.084967320261438, "step": 6250}, {"loss": 1.0977, "grad_norm": 0.8343415856361389, "learning_rate": 0.0002, "epoch": 4.091503267973856, "step": 6260}, {"loss": 0.9887, "grad_norm": 0.7490310072898865, "learning_rate": 0.0002, "epoch": 4.098039215686274, "step": 6270}, {"loss": 1.2084, "grad_norm": 0.8961182832717896, "learning_rate": 0.0002, "epoch": 4.104575163398692, "step": 6280}, {"loss": 1.1349, "grad_norm": 0.7124854922294617, "learning_rate": 0.0002, "epoch": 4.111111111111111, "step": 6290}, {"loss": 1.0081, "grad_norm": 0.8338138461112976, "learning_rate": 0.0002, "epoch": 4.117647058823529, "step": 6300}, {"loss": 1.1091, "grad_norm": 0.8075833320617676, "learning_rate": 0.0002, "epoch": 4.124183006535947, "step": 6310}, {"loss": 1.0193, "grad_norm": 0.8069391846656799, "learning_rate": 0.0002, "epoch": 4.130718954248366, "step": 6320}, {"loss": 0.948, "grad_norm": 0.9567893147468567, "learning_rate": 0.0002, "epoch": 4.137254901960785, "step": 6330}, {"loss": 1.0241, "grad_norm": 1.2184662818908691, "learning_rate": 0.0002, "epoch": 4.143790849673203, "step": 6340}, {"loss": 1.0756, "grad_norm": 1.030976414680481, "learning_rate": 0.0002, "epoch": 4.150326797385621, "step": 6350}, {"loss": 1.1124, "grad_norm": 0.9749957323074341, "learning_rate": 0.0002, "epoch": 4.1568627450980395, "step": 6360}, {"loss": 1.1038, "grad_norm": 0.7089483141899109, "learning_rate": 0.0002, "epoch": 4.163398692810458, "step": 6370}, {"loss": 1.2175, "grad_norm": 1.1084946393966675, "learning_rate": 0.0002, "epoch": 4.169934640522876, "step": 6380}, {"loss": 1.0274, "grad_norm": 0.7998497486114502, "learning_rate": 0.0002, "epoch": 4.176470588235294, "step": 6390}, {"loss": 1.005, "grad_norm": 0.8997811675071716, "learning_rate": 0.0002, "epoch": 4.183006535947713, "step": 6400}, {"loss": 1.0704, "grad_norm": 0.8359479904174805, "learning_rate": 0.0002, "epoch": 4.189542483660131, "step": 6410}, {"loss": 1.1056, "grad_norm": 0.9087472558021545, "learning_rate": 0.0002, "epoch": 4.196078431372549, "step": 6420}, {"loss": 1.0657, "grad_norm": 1.1100451946258545, "learning_rate": 0.0002, "epoch": 4.2026143790849675, "step": 6430}, {"loss": 1.1443, "grad_norm": 0.9376999735832214, "learning_rate": 0.0002, "epoch": 4.209150326797386, "step": 6440}, {"loss": 1.0862, "grad_norm": 0.8179266452789307, "learning_rate": 0.0002, "epoch": 4.215686274509804, "step": 6450}, {"loss": 1.0679, "grad_norm": 0.9953271746635437, "learning_rate": 0.0002, "epoch": 4.222222222222222, "step": 6460}, {"loss": 1.1034, "grad_norm": 0.8476650714874268, "learning_rate": 0.0002, "epoch": 4.228758169934641, "step": 6470}, {"loss": 1.2512, "grad_norm": 0.8406323194503784, "learning_rate": 0.0002, "epoch": 4.235294117647059, "step": 6480}, {"loss": 1.057, "grad_norm": 0.819134533405304, "learning_rate": 0.0002, "epoch": 4.241830065359477, "step": 6490}, {"loss": 1.1082, "grad_norm": 0.7764983773231506, "learning_rate": 0.0002, "epoch": 4.248366013071895, "step": 6500}, {"loss": 1.1593, "grad_norm": 0.8252112865447998, "learning_rate": 0.0002, "epoch": 4.254901960784314, "step": 6510}, {"loss": 1.1369, "grad_norm": 0.7941019535064697, "learning_rate": 0.0002, "epoch": 4.261437908496732, "step": 6520}, {"loss": 1.0296, "grad_norm": 0.7673905491828918, "learning_rate": 0.0002, "epoch": 4.26797385620915, "step": 6530}, {"loss": 1.1387, "grad_norm": 0.8749890327453613, "learning_rate": 0.0002, "epoch": 4.2745098039215685, "step": 6540}, {"loss": 1.0595, "grad_norm": 0.7343207597732544, "learning_rate": 0.0002, "epoch": 4.281045751633987, "step": 6550}, {"loss": 1.1715, "grad_norm": 1.2786651849746704, "learning_rate": 0.0002, "epoch": 4.287581699346405, "step": 6560}, {"loss": 1.0514, "grad_norm": 1.316875696182251, "learning_rate": 0.0002, "epoch": 4.294117647058823, "step": 6570}, {"loss": 1.1125, "grad_norm": 0.8349189162254333, "learning_rate": 0.0002, "epoch": 4.300653594771242, "step": 6580}, {"loss": 1.0732, "grad_norm": 0.7510647177696228, "learning_rate": 0.0002, "epoch": 4.30718954248366, "step": 6590}, {"loss": 1.1387, "grad_norm": 0.932420551776886, "learning_rate": 0.0002, "epoch": 4.313725490196078, "step": 6600}, {"loss": 1.1115, "grad_norm": 0.8510616421699524, "learning_rate": 0.0002, "epoch": 4.3202614379084965, "step": 6610}, {"loss": 1.0957, "grad_norm": 0.7661547064781189, "learning_rate": 0.0002, "epoch": 4.326797385620915, "step": 6620}, {"loss": 1.2064, "grad_norm": 1.0370930433273315, "learning_rate": 0.0002, "epoch": 4.333333333333333, "step": 6630}, {"loss": 1.1064, "grad_norm": 0.9302158951759338, "learning_rate": 0.0002, "epoch": 4.339869281045751, "step": 6640}, {"loss": 0.968, "grad_norm": 0.9203811883926392, "learning_rate": 0.0002, "epoch": 4.34640522875817, "step": 6650}, {"loss": 1.0123, "grad_norm": 0.9986332654953003, "learning_rate": 0.0002, "epoch": 4.352941176470588, "step": 6660}, {"loss": 1.1079, "grad_norm": 0.8001713156700134, "learning_rate": 0.0002, "epoch": 4.359477124183006, "step": 6670}, {"loss": 1.0248, "grad_norm": 0.829714298248291, "learning_rate": 0.0002, "epoch": 4.366013071895424, "step": 6680}, {"loss": 1.0389, "grad_norm": 0.8253079056739807, "learning_rate": 0.0002, "epoch": 4.372549019607844, "step": 6690}, {"loss": 1.1087, "grad_norm": 0.824666440486908, "learning_rate": 0.0002, "epoch": 4.379084967320262, "step": 6700}, {"loss": 1.1968, "grad_norm": 0.8872972130775452, "learning_rate": 0.0002, "epoch": 4.38562091503268, "step": 6710}, {"loss": 1.0474, "grad_norm": 0.8729761838912964, "learning_rate": 0.0002, "epoch": 4.392156862745098, "step": 6720}, {"loss": 1.0961, "grad_norm": 1.1367264986038208, "learning_rate": 0.0002, "epoch": 4.398692810457517, "step": 6730}, {"loss": 1.0184, "grad_norm": 0.9699058532714844, "learning_rate": 0.0002, "epoch": 4.405228758169935, "step": 6740}, {"loss": 1.006, "grad_norm": 0.8266763687133789, "learning_rate": 0.0002, "epoch": 4.411764705882353, "step": 6750}, {"loss": 1.0735, "grad_norm": 1.0249767303466797, "learning_rate": 0.0002, "epoch": 4.4183006535947715, "step": 6760}, {"loss": 1.1726, "grad_norm": 0.73606938123703, "learning_rate": 0.0002, "epoch": 4.42483660130719, "step": 6770}, {"loss": 1.1037, "grad_norm": 1.4050679206848145, "learning_rate": 0.0002, "epoch": 4.431372549019608, "step": 6780}, {"loss": 1.1418, "grad_norm": 1.1114081144332886, "learning_rate": 0.0002, "epoch": 4.437908496732026, "step": 6790}, {"loss": 0.9682, "grad_norm": 0.8031067848205566, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6800}, {"loss": 1.0753, "grad_norm": 0.8513566851615906, "learning_rate": 0.0002, "epoch": 4.450980392156863, "step": 6810}, {"loss": 1.1852, "grad_norm": 1.332741379737854, "learning_rate": 0.0002, "epoch": 4.457516339869281, "step": 6820}, {"loss": 1.0966, "grad_norm": 1.5032578706741333, "learning_rate": 0.0002, "epoch": 4.4640522875816995, "step": 6830}, {"loss": 1.1124, "grad_norm": 0.7677283883094788, "learning_rate": 0.0002, "epoch": 4.470588235294118, "step": 6840}, {"loss": 1.1501, "grad_norm": 0.989148736000061, "learning_rate": 0.0002, "epoch": 4.477124183006536, "step": 6850}, {"loss": 1.2239, "grad_norm": 1.5316275358200073, "learning_rate": 0.0002, "epoch": 4.483660130718954, "step": 6860}, {"loss": 1.1171, "grad_norm": 0.9427124261856079, "learning_rate": 0.0002, "epoch": 4.490196078431373, "step": 6870}, {"loss": 1.1314, "grad_norm": 1.215287685394287, "learning_rate": 0.0002, "epoch": 4.496732026143791, "step": 6880}, {"loss": 1.0809, "grad_norm": 0.7286760210990906, "learning_rate": 0.0002, "epoch": 4.503267973856209, "step": 6890}, {"loss": 1.0179, "grad_norm": 0.874829888343811, "learning_rate": 0.0002, "epoch": 4.509803921568627, "step": 6900}, {"loss": 1.0233, "grad_norm": 0.8058359622955322, "learning_rate": 0.0002, "epoch": 4.516339869281046, "step": 6910}, {"loss": 1.0463, "grad_norm": 1.248195767402649, "learning_rate": 0.0002, "epoch": 4.522875816993464, "step": 6920}, {"loss": 1.0347, "grad_norm": 0.8033645749092102, "learning_rate": 0.0002, "epoch": 4.529411764705882, "step": 6930}, {"loss": 1.1068, "grad_norm": 1.7361950874328613, "learning_rate": 0.0002, "epoch": 4.5359477124183005, "step": 6940}, {"loss": 0.9856, "grad_norm": 0.8058095574378967, "learning_rate": 0.0002, "epoch": 4.542483660130719, "step": 6950}, {"loss": 1.0057, "grad_norm": 1.254089593887329, "learning_rate": 0.0002, "epoch": 4.549019607843137, "step": 6960}, {"loss": 1.1723, "grad_norm": 0.9180455803871155, "learning_rate": 0.0002, "epoch": 4.555555555555555, "step": 6970}, {"loss": 1.0559, "grad_norm": 0.6677682399749756, "learning_rate": 0.0002, "epoch": 4.562091503267974, "step": 6980}, {"loss": 1.0453, "grad_norm": 0.8127354383468628, "learning_rate": 0.0002, "epoch": 4.568627450980392, "step": 6990}, {"loss": 1.0828, "grad_norm": 1.0263001918792725, "learning_rate": 0.0002, "epoch": 4.57516339869281, "step": 7000}, {"loss": 1.0703, "grad_norm": 0.9641909003257751, "learning_rate": 0.0002, "epoch": 4.5816993464052285, "step": 7010}, {"loss": 1.179, "grad_norm": 0.9440861344337463, "learning_rate": 0.0002, "epoch": 4.588235294117647, "step": 7020}, {"loss": 1.0931, "grad_norm": 0.9539011716842651, "learning_rate": 0.0002, "epoch": 4.594771241830065, "step": 7030}, {"loss": 1.0963, "grad_norm": 1.0449910163879395, "learning_rate": 0.0002, "epoch": 4.601307189542483, "step": 7040}, {"loss": 0.9944, "grad_norm": 0.8766893744468689, "learning_rate": 0.0002, "epoch": 4.607843137254902, "step": 7050}, {"loss": 1.0169, "grad_norm": 0.6983462572097778, "learning_rate": 0.0002, "epoch": 4.61437908496732, "step": 7060}, {"loss": 1.1778, "grad_norm": 0.9505505561828613, "learning_rate": 0.0002, "epoch": 4.620915032679738, "step": 7070}, {"loss": 1.121, "grad_norm": 1.2506657838821411, "learning_rate": 0.0002, "epoch": 4.627450980392156, "step": 7080}, {"loss": 1.1329, "grad_norm": 0.9602801203727722, "learning_rate": 0.0002, "epoch": 4.633986928104575, "step": 7090}, {"loss": 1.1499, "grad_norm": 0.7398977875709534, "learning_rate": 0.0002, "epoch": 4.640522875816993, "step": 7100}, {"loss": 1.0769, "grad_norm": 1.3862425088882446, "learning_rate": 0.0002, "epoch": 4.647058823529412, "step": 7110}, {"loss": 1.0571, "grad_norm": 1.1451990604400635, "learning_rate": 0.0002, "epoch": 4.65359477124183, "step": 7120}, {"loss": 1.1271, "grad_norm": 0.9010422229766846, "learning_rate": 0.0002, "epoch": 4.660130718954249, "step": 7130}, {"loss": 1.0165, "grad_norm": 0.7102518081665039, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 7140}, {"loss": 1.0819, "grad_norm": 0.7963796257972717, "learning_rate": 0.0002, "epoch": 4.673202614379085, "step": 7150}, {"loss": 1.1114, "grad_norm": 0.7726007699966431, "learning_rate": 0.0002, "epoch": 4.6797385620915035, "step": 7160}, {"loss": 1.2088, "grad_norm": 0.8097564578056335, "learning_rate": 0.0002, "epoch": 4.686274509803922, "step": 7170}, {"loss": 1.1386, "grad_norm": 0.9070925116539001, "learning_rate": 0.0002, "epoch": 4.69281045751634, "step": 7180}, {"loss": 1.0315, "grad_norm": 0.7543528079986572, "learning_rate": 0.0002, "epoch": 4.699346405228758, "step": 7190}, {"loss": 1.0984, "grad_norm": 0.9900904893875122, "learning_rate": 0.0002, "epoch": 4.705882352941177, "step": 7200}, {"loss": 1.1552, "grad_norm": 0.8033412098884583, "learning_rate": 0.0002, "epoch": 4.712418300653595, "step": 7210}, {"loss": 1.1773, "grad_norm": 0.8440839052200317, "learning_rate": 0.0002, "epoch": 4.718954248366013, "step": 7220}, {"loss": 1.1258, "grad_norm": 0.9325555562973022, "learning_rate": 0.0002, "epoch": 4.7254901960784315, "step": 7230}, {"loss": 1.1384, "grad_norm": 0.7881146669387817, "learning_rate": 0.0002, "epoch": 4.73202614379085, "step": 7240}, {"loss": 1.1219, "grad_norm": 0.884453296661377, "learning_rate": 0.0002, "epoch": 4.738562091503268, "step": 7250}, {"loss": 1.1036, "grad_norm": 0.9274539351463318, "learning_rate": 0.0002, "epoch": 4.745098039215686, "step": 7260}, {"loss": 1.0906, "grad_norm": 1.2367479801177979, "learning_rate": 0.0002, "epoch": 4.751633986928105, "step": 7270}, {"loss": 1.0741, "grad_norm": 0.9499821066856384, "learning_rate": 0.0002, "epoch": 4.758169934640523, "step": 7280}, {"loss": 1.1625, "grad_norm": 2.1918580532073975, "learning_rate": 0.0002, "epoch": 4.764705882352941, "step": 7290}, {"loss": 0.954, "grad_norm": 0.8221880793571472, "learning_rate": 0.0002, "epoch": 4.771241830065359, "step": 7300}, {"loss": 1.1358, "grad_norm": 0.871972918510437, "learning_rate": 0.0002, "epoch": 4.777777777777778, "step": 7310}, {"loss": 1.0599, "grad_norm": 0.8034510612487793, "learning_rate": 0.0002, "epoch": 4.784313725490196, "step": 7320}, {"loss": 1.1059, "grad_norm": 0.8959605693817139, "learning_rate": 0.0002, "epoch": 4.790849673202614, "step": 7330}, {"loss": 1.0176, "grad_norm": 1.2326215505599976, "learning_rate": 0.0002, "epoch": 4.7973856209150325, "step": 7340}, {"loss": 1.1095, "grad_norm": 0.9725791811943054, "learning_rate": 0.0002, "epoch": 4.803921568627451, "step": 7350}, {"loss": 1.1229, "grad_norm": 0.7240816354751587, "learning_rate": 0.0002, "epoch": 4.810457516339869, "step": 7360}, {"loss": 1.0669, "grad_norm": 0.8265769481658936, "learning_rate": 0.0002, "epoch": 4.816993464052287, "step": 7370}, {"loss": 1.042, "grad_norm": 0.8888696432113647, "learning_rate": 0.0002, "epoch": 4.823529411764706, "step": 7380}, {"loss": 1.0981, "grad_norm": 0.7776556015014648, "learning_rate": 0.0002, "epoch": 4.830065359477124, "step": 7390}, {"loss": 1.0819, "grad_norm": 0.8772371411323547, "learning_rate": 0.0002, "epoch": 4.836601307189542, "step": 7400}, {"loss": 1.0819, "grad_norm": 0.9786531925201416, "learning_rate": 0.0002, "epoch": 4.8431372549019605, "step": 7410}, {"loss": 1.1358, "grad_norm": 0.9059745073318481, "learning_rate": 0.0002, "epoch": 4.849673202614379, "step": 7420}, {"loss": 1.0324, "grad_norm": 0.7422552108764648, "learning_rate": 0.0002, "epoch": 4.856209150326797, "step": 7430}, {"loss": 1.0423, "grad_norm": 1.3040380477905273, "learning_rate": 0.0002, "epoch": 4.862745098039216, "step": 7440}, {"loss": 1.1161, "grad_norm": 1.3278473615646362, "learning_rate": 0.0002, "epoch": 4.8692810457516345, "step": 7450}, {"loss": 1.0713, "grad_norm": 1.2705849409103394, "learning_rate": 0.0002, "epoch": 4.875816993464053, "step": 7460}, {"loss": 1.0034, "grad_norm": 0.8837892413139343, "learning_rate": 0.0002, "epoch": 4.882352941176471, "step": 7470}, {"loss": 1.1716, "grad_norm": 0.8670691251754761, "learning_rate": 0.0002, "epoch": 4.888888888888889, "step": 7480}, {"loss": 1.1723, "grad_norm": 0.9662758111953735, "learning_rate": 0.0002, "epoch": 4.895424836601308, "step": 7490}, {"loss": 1.1056, "grad_norm": 0.8188302516937256, "learning_rate": 0.0002, "epoch": 4.901960784313726, "step": 7500}, {"loss": 1.0419, "grad_norm": 0.769442617893219, "learning_rate": 0.0002, "epoch": 4.908496732026144, "step": 7510}, {"loss": 1.1671, "grad_norm": 1.1465084552764893, "learning_rate": 0.0002, "epoch": 4.915032679738562, "step": 7520}, {"loss": 1.0768, "grad_norm": 1.253214955329895, "learning_rate": 0.0002, "epoch": 4.921568627450981, "step": 7530}, {"loss": 1.011, "grad_norm": 0.7922375202178955, "learning_rate": 0.0002, "epoch": 4.928104575163399, "step": 7540}, {"loss": 1.1256, "grad_norm": 0.8306851387023926, "learning_rate": 0.0002, "epoch": 4.934640522875817, "step": 7550}, {"loss": 1.206, "grad_norm": 0.8486151099205017, "learning_rate": 0.0002, "epoch": 4.9411764705882355, "step": 7560}, {"loss": 1.0161, "grad_norm": 1.2601467370986938, "learning_rate": 0.0002, "epoch": 4.947712418300654, "step": 7570}, {"loss": 1.1078, "grad_norm": 0.7980747818946838, "learning_rate": 0.0002, "epoch": 4.954248366013072, "step": 7580}, {"loss": 1.0607, "grad_norm": 0.8653254508972168, "learning_rate": 0.0002, "epoch": 4.96078431372549, "step": 7590}, {"loss": 1.0292, "grad_norm": 0.9680571556091309, "learning_rate": 0.0002, "epoch": 4.967320261437909, "step": 7600}, {"loss": 1.1795, "grad_norm": 0.9554466605186462, "learning_rate": 0.0002, "epoch": 4.973856209150327, "step": 7610}, {"loss": 1.0935, "grad_norm": 1.3693897724151611, "learning_rate": 0.0002, "epoch": 4.980392156862745, "step": 7620}, {"loss": 1.0838, "grad_norm": 0.7809282541275024, "learning_rate": 0.0002, "epoch": 4.9869281045751634, "step": 7630}, {"loss": 1.0844, "grad_norm": 0.7528006434440613, "learning_rate": 0.0002, "epoch": 4.993464052287582, "step": 7640}, {"loss": 0.9951, "grad_norm": 1.7491309642791748, "learning_rate": 0.0002, "epoch": 5.0, "step": 7650}, {"eval_loss": 1.4197258949279785, "eval_runtime": 33.6327, "eval_samples_per_second": 12.964, "eval_steps_per_second": 1.635, "epoch": 5.0, "step": 7650}, {"loss": 0.9744, "grad_norm": 0.8840063214302063, "learning_rate": 0.0002, "epoch": 5.006535947712418, "step": 7660}, {"loss": 1.0274, "grad_norm": 1.0118401050567627, "learning_rate": 0.0002, "epoch": 5.0130718954248366, "step": 7670}, {"loss": 1.1667, "grad_norm": 1.0040518045425415, "learning_rate": 0.0002, "epoch": 5.019607843137255, "step": 7680}, {"loss": 0.9426, "grad_norm": 0.7541199922561646, "learning_rate": 0.0002, "epoch": 5.026143790849673, "step": 7690}, {"loss": 1.0797, "grad_norm": 0.9106482863426208, "learning_rate": 0.0002, "epoch": 5.032679738562091, "step": 7700}, {"loss": 1.0096, "grad_norm": 1.3691469430923462, "learning_rate": 0.0002, "epoch": 5.03921568627451, "step": 7710}, {"loss": 0.9889, "grad_norm": 0.9449689388275146, "learning_rate": 0.0002, "epoch": 5.045751633986928, "step": 7720}, {"loss": 0.9087, "grad_norm": 1.1678508520126343, "learning_rate": 0.0002, "epoch": 5.052287581699346, "step": 7730}, {"loss": 1.0556, "grad_norm": 1.1296145915985107, "learning_rate": 0.0002, "epoch": 5.0588235294117645, "step": 7740}, {"loss": 0.9339, "grad_norm": 0.7863904237747192, "learning_rate": 0.0002, "epoch": 5.065359477124183, "step": 7750}, {"loss": 1.0135, "grad_norm": 0.8691433072090149, "learning_rate": 0.0002, "epoch": 5.071895424836601, "step": 7760}, {"loss": 0.9776, "grad_norm": 1.0722088813781738, "learning_rate": 0.0002, "epoch": 5.078431372549019, "step": 7770}, {"loss": 1.0595, "grad_norm": 0.9625038504600525, "learning_rate": 0.0002, "epoch": 5.084967320261438, "step": 7780}, {"loss": 1.0241, "grad_norm": 1.2618783712387085, "learning_rate": 0.0002, "epoch": 5.091503267973856, "step": 7790}, {"loss": 0.9396, "grad_norm": 0.9970650672912598, "learning_rate": 0.0002, "epoch": 5.098039215686274, "step": 7800}, {"loss": 0.9186, "grad_norm": 1.3946677446365356, "learning_rate": 0.0002, "epoch": 5.104575163398692, "step": 7810}, {"loss": 0.9957, "grad_norm": 1.0260052680969238, "learning_rate": 0.0002, "epoch": 5.111111111111111, "step": 7820}, {"loss": 0.9865, "grad_norm": 1.105521559715271, "learning_rate": 0.0002, "epoch": 5.117647058823529, "step": 7830}, {"loss": 0.9788, "grad_norm": 1.003641128540039, "learning_rate": 0.0002, "epoch": 5.124183006535947, "step": 7840}, {"loss": 0.9688, "grad_norm": 1.0315021276474, "learning_rate": 0.0002, "epoch": 5.130718954248366, "step": 7850}, {"loss": 1.0001, "grad_norm": 0.9469530582427979, "learning_rate": 0.0002, "epoch": 5.137254901960785, "step": 7860}, {"loss": 0.9659, "grad_norm": 1.3244667053222656, "learning_rate": 0.0002, "epoch": 5.143790849673203, "step": 7870}, {"loss": 0.9657, "grad_norm": 1.1732033491134644, "learning_rate": 0.0002, "epoch": 5.150326797385621, "step": 7880}, {"loss": 0.9978, "grad_norm": 1.3129149675369263, "learning_rate": 0.0002, "epoch": 5.1568627450980395, "step": 7890}, {"loss": 0.9894, "grad_norm": 0.8589454293251038, "learning_rate": 0.0002, "epoch": 5.163398692810458, "step": 7900}, {"loss": 1.0161, "grad_norm": 0.8954233527183533, "learning_rate": 0.0002, "epoch": 5.169934640522876, "step": 7910}, {"loss": 0.8741, "grad_norm": 0.7426522970199585, "learning_rate": 0.0002, "epoch": 5.176470588235294, "step": 7920}, {"loss": 1.0106, "grad_norm": 1.1990121603012085, "learning_rate": 0.0002, "epoch": 5.183006535947713, "step": 7930}, {"loss": 0.9453, "grad_norm": 0.8867580890655518, "learning_rate": 0.0002, "epoch": 5.189542483660131, "step": 7940}, {"loss": 0.9727, "grad_norm": 1.016276478767395, "learning_rate": 0.0002, "epoch": 5.196078431372549, "step": 7950}, {"loss": 0.9908, "grad_norm": 1.0210685729980469, "learning_rate": 0.0002, "epoch": 5.2026143790849675, "step": 7960}, {"loss": 1.0522, "grad_norm": 1.0093122720718384, "learning_rate": 0.0002, "epoch": 5.209150326797386, "step": 7970}, {"loss": 1.0055, "grad_norm": 0.9746801853179932, "learning_rate": 0.0002, "epoch": 5.215686274509804, "step": 7980}, {"loss": 1.0611, "grad_norm": 0.9113537073135376, "learning_rate": 0.0002, "epoch": 5.222222222222222, "step": 7990}, {"loss": 0.9167, "grad_norm": 1.2782206535339355, "learning_rate": 0.0002, "epoch": 5.228758169934641, "step": 8000}, {"loss": 1.0212, "grad_norm": 1.3223118782043457, "learning_rate": 0.0002, "epoch": 5.235294117647059, "step": 8010}, {"loss": 0.9244, "grad_norm": 0.7898629307746887, "learning_rate": 0.0002, "epoch": 5.241830065359477, "step": 8020}, {"loss": 1.0574, "grad_norm": 0.9822350740432739, "learning_rate": 0.0002, "epoch": 5.248366013071895, "step": 8030}, {"loss": 1.0102, "grad_norm": 1.5114340782165527, "learning_rate": 0.0002, "epoch": 5.254901960784314, "step": 8040}, {"loss": 0.9816, "grad_norm": 0.859006941318512, "learning_rate": 0.0002, "epoch": 5.261437908496732, "step": 8050}, {"loss": 0.9445, "grad_norm": 1.0495043992996216, "learning_rate": 0.0002, "epoch": 5.26797385620915, "step": 8060}, {"loss": 0.9724, "grad_norm": 1.329483151435852, "learning_rate": 0.0002, "epoch": 5.2745098039215685, "step": 8070}, {"loss": 0.9296, "grad_norm": 1.1333061456680298, "learning_rate": 0.0002, "epoch": 5.281045751633987, "step": 8080}, {"loss": 0.9577, "grad_norm": 0.8153108358383179, "learning_rate": 0.0002, "epoch": 5.287581699346405, "step": 8090}, {"loss": 0.9002, "grad_norm": 0.9395004510879517, "learning_rate": 0.0002, "epoch": 5.294117647058823, "step": 8100}, {"loss": 1.0371, "grad_norm": 0.8907593488693237, "learning_rate": 0.0002, "epoch": 5.300653594771242, "step": 8110}, {"loss": 0.9301, "grad_norm": 0.9808667898178101, "learning_rate": 0.0002, "epoch": 5.30718954248366, "step": 8120}, {"loss": 1.0136, "grad_norm": 0.984779417514801, "learning_rate": 0.0002, "epoch": 5.313725490196078, "step": 8130}, {"loss": 0.9621, "grad_norm": 0.9787270426750183, "learning_rate": 0.0002, "epoch": 5.3202614379084965, "step": 8140}, {"loss": 0.9336, "grad_norm": 0.9857710599899292, "learning_rate": 0.0002, "epoch": 5.326797385620915, "step": 8150}, {"loss": 0.9884, "grad_norm": 0.9774303436279297, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 8160}, {"loss": 1.0561, "grad_norm": 0.677925169467926, "learning_rate": 0.0002, "epoch": 5.339869281045751, "step": 8170}, {"loss": 1.1345, "grad_norm": 0.9576456546783447, "learning_rate": 0.0002, "epoch": 5.34640522875817, "step": 8180}, {"loss": 0.9554, "grad_norm": 1.8970937728881836, "learning_rate": 0.0002, "epoch": 5.352941176470588, "step": 8190}, {"loss": 1.0474, "grad_norm": 0.9458389282226562, "learning_rate": 0.0002, "epoch": 5.359477124183006, "step": 8200}, {"loss": 1.0365, "grad_norm": 1.761794924736023, "learning_rate": 0.0002, "epoch": 5.366013071895424, "step": 8210}, {"loss": 0.9426, "grad_norm": 1.0693724155426025, "learning_rate": 0.0002, "epoch": 5.372549019607844, "step": 8220}, {"loss": 1.0299, "grad_norm": 0.9025877714157104, "learning_rate": 0.0002, "epoch": 5.379084967320262, "step": 8230}, {"loss": 0.9652, "grad_norm": 1.258857250213623, "learning_rate": 0.0002, "epoch": 5.38562091503268, "step": 8240}, {"loss": 0.9735, "grad_norm": 1.084849238395691, "learning_rate": 0.0002, "epoch": 5.392156862745098, "step": 8250}, {"loss": 0.9999, "grad_norm": 0.9530340433120728, "learning_rate": 0.0002, "epoch": 5.398692810457517, "step": 8260}, {"loss": 1.0268, "grad_norm": 0.830240786075592, "learning_rate": 0.0002, "epoch": 5.405228758169935, "step": 8270}, {"loss": 1.0332, "grad_norm": 1.5807015895843506, "learning_rate": 0.0002, "epoch": 5.411764705882353, "step": 8280}, {"loss": 0.9146, "grad_norm": 0.9486905336380005, "learning_rate": 0.0002, "epoch": 5.4183006535947715, "step": 8290}, {"loss": 1.0336, "grad_norm": 1.0415093898773193, "learning_rate": 0.0002, "epoch": 5.42483660130719, "step": 8300}, {"loss": 0.8933, "grad_norm": 1.0501102209091187, "learning_rate": 0.0002, "epoch": 5.431372549019608, "step": 8310}, {"loss": 0.9983, "grad_norm": 0.9751836061477661, "learning_rate": 0.0002, "epoch": 5.437908496732026, "step": 8320}, {"loss": 1.0755, "grad_norm": 1.5529173612594604, "learning_rate": 0.0002, "epoch": 5.444444444444445, "step": 8330}, {"loss": 0.9814, "grad_norm": 0.8314350247383118, "learning_rate": 0.0002, "epoch": 5.450980392156863, "step": 8340}, {"loss": 1.0596, "grad_norm": 1.2555103302001953, "learning_rate": 0.0002, "epoch": 5.457516339869281, "step": 8350}, {"loss": 1.0127, "grad_norm": 0.9408367872238159, "learning_rate": 0.0002, "epoch": 5.4640522875816995, "step": 8360}, {"loss": 0.9241, "grad_norm": 0.9483312964439392, "learning_rate": 0.0002, "epoch": 5.470588235294118, "step": 8370}, {"loss": 0.9678, "grad_norm": 0.957905650138855, "learning_rate": 0.0002, "epoch": 5.477124183006536, "step": 8380}, {"loss": 1.0985, "grad_norm": 1.4000147581100464, "learning_rate": 0.0002, "epoch": 5.483660130718954, "step": 8390}, {"loss": 0.9966, "grad_norm": 1.7032461166381836, "learning_rate": 0.0002, "epoch": 5.490196078431373, "step": 8400}, {"loss": 0.9539, "grad_norm": 0.8978716731071472, "learning_rate": 0.0002, "epoch": 5.496732026143791, "step": 8410}, {"loss": 0.9544, "grad_norm": 0.8659300804138184, "learning_rate": 0.0002, "epoch": 5.503267973856209, "step": 8420}, {"loss": 1.0526, "grad_norm": 1.3629727363586426, "learning_rate": 0.0002, "epoch": 5.509803921568627, "step": 8430}, {"loss": 0.9696, "grad_norm": 1.2741984128952026, "learning_rate": 0.0002, "epoch": 5.516339869281046, "step": 8440}, {"loss": 1.0191, "grad_norm": 1.3867180347442627, "learning_rate": 0.0002, "epoch": 5.522875816993464, "step": 8450}, {"loss": 1.0835, "grad_norm": 1.0662001371383667, "learning_rate": 0.0002, "epoch": 5.529411764705882, "step": 8460}, {"loss": 0.9779, "grad_norm": 1.7005380392074585, "learning_rate": 0.0002, "epoch": 5.5359477124183005, "step": 8470}, {"loss": 1.0221, "grad_norm": 1.3730385303497314, "learning_rate": 0.0002, "epoch": 5.542483660130719, "step": 8480}, {"loss": 0.9586, "grad_norm": 1.7737441062927246, "learning_rate": 0.0002, "epoch": 5.549019607843137, "step": 8490}, {"loss": 0.9729, "grad_norm": 0.907487690448761, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 8500}, {"loss": 0.9891, "grad_norm": 0.8882441520690918, "learning_rate": 0.0002, "epoch": 5.562091503267974, "step": 8510}, {"loss": 0.973, "grad_norm": 0.8655388951301575, "learning_rate": 0.0002, "epoch": 5.568627450980392, "step": 8520}, {"loss": 0.9523, "grad_norm": 1.379992961883545, "learning_rate": 0.0002, "epoch": 5.57516339869281, "step": 8530}, {"loss": 1.0174, "grad_norm": 1.0021201372146606, "learning_rate": 0.0002, "epoch": 5.5816993464052285, "step": 8540}, {"loss": 1.0113, "grad_norm": 1.2636926174163818, "learning_rate": 0.0002, "epoch": 5.588235294117647, "step": 8550}, {"loss": 1.0243, "grad_norm": 1.279025912284851, "learning_rate": 0.0002, "epoch": 5.594771241830065, "step": 8560}, {"loss": 0.9917, "grad_norm": 0.8885834217071533, "learning_rate": 0.0002, "epoch": 5.601307189542483, "step": 8570}, {"loss": 0.9849, "grad_norm": 1.1975032091140747, "learning_rate": 0.0002, "epoch": 5.607843137254902, "step": 8580}, {"loss": 1.0363, "grad_norm": 1.005470871925354, "learning_rate": 0.0002, "epoch": 5.61437908496732, "step": 8590}, {"loss": 0.9947, "grad_norm": 1.104286551475525, "learning_rate": 0.0002, "epoch": 5.620915032679738, "step": 8600}, {"loss": 1.0585, "grad_norm": 1.435445785522461, "learning_rate": 0.0002, "epoch": 5.627450980392156, "step": 8610}, {"loss": 0.9156, "grad_norm": 1.0270172357559204, "learning_rate": 0.0002, "epoch": 5.633986928104575, "step": 8620}, {"loss": 1.0522, "grad_norm": 1.0929527282714844, "learning_rate": 0.0002, "epoch": 5.640522875816993, "step": 8630}, {"loss": 0.9694, "grad_norm": 1.1061221361160278, "learning_rate": 0.0002, "epoch": 5.647058823529412, "step": 8640}, {"loss": 1.0826, "grad_norm": 0.9563149213790894, "learning_rate": 0.0002, "epoch": 5.65359477124183, "step": 8650}, {"loss": 1.0042, "grad_norm": 1.0434954166412354, "learning_rate": 0.0002, "epoch": 5.660130718954249, "step": 8660}, {"loss": 0.9463, "grad_norm": 1.3695117235183716, "learning_rate": 0.0002, "epoch": 5.666666666666667, "step": 8670}, {"loss": 0.9441, "grad_norm": 1.0540564060211182, "learning_rate": 0.0002, "epoch": 5.673202614379085, "step": 8680}, {"loss": 0.9755, "grad_norm": 1.5942492485046387, "learning_rate": 0.0002, "epoch": 5.6797385620915035, "step": 8690}, {"loss": 1.0071, "grad_norm": 0.9485495090484619, "learning_rate": 0.0002, "epoch": 5.686274509803922, "step": 8700}, {"loss": 0.9998, "grad_norm": 1.1483162641525269, "learning_rate": 0.0002, "epoch": 5.69281045751634, "step": 8710}, {"loss": 0.9578, "grad_norm": 0.9075471758842468, "learning_rate": 0.0002, "epoch": 5.699346405228758, "step": 8720}, {"loss": 0.9488, "grad_norm": 1.7908551692962646, "learning_rate": 0.0002, "epoch": 5.705882352941177, "step": 8730}, {"loss": 1.0163, "grad_norm": 0.8867162466049194, "learning_rate": 0.0002, "epoch": 5.712418300653595, "step": 8740}, {"loss": 1.0041, "grad_norm": 1.7165148258209229, "learning_rate": 0.0002, "epoch": 5.718954248366013, "step": 8750}, {"loss": 1.1061, "grad_norm": 0.9529356956481934, "learning_rate": 0.0002, "epoch": 5.7254901960784315, "step": 8760}, {"loss": 1.1119, "grad_norm": 1.01852548122406, "learning_rate": 0.0002, "epoch": 5.73202614379085, "step": 8770}, {"loss": 1.0471, "grad_norm": 0.9538423418998718, "learning_rate": 0.0002, "epoch": 5.738562091503268, "step": 8780}, {"loss": 1.0913, "grad_norm": 0.9007737636566162, "learning_rate": 0.0002, "epoch": 5.745098039215686, "step": 8790}, {"loss": 0.9766, "grad_norm": 0.9107874035835266, "learning_rate": 0.0002, "epoch": 5.751633986928105, "step": 8800}, {"loss": 0.9212, "grad_norm": 0.7379238605499268, "learning_rate": 0.0002, "epoch": 5.758169934640523, "step": 8810}, {"loss": 1.0966, "grad_norm": 1.072645902633667, "learning_rate": 0.0002, "epoch": 5.764705882352941, "step": 8820}, {"loss": 1.0845, "grad_norm": 1.002008080482483, "learning_rate": 0.0002, "epoch": 5.771241830065359, "step": 8830}, {"loss": 0.9978, "grad_norm": 1.0435924530029297, "learning_rate": 0.0002, "epoch": 5.777777777777778, "step": 8840}, {"loss": 0.9458, "grad_norm": 0.9874551296234131, "learning_rate": 0.0002, "epoch": 5.784313725490196, "step": 8850}, {"loss": 1.1241, "grad_norm": 1.1729662418365479, "learning_rate": 0.0002, "epoch": 5.790849673202614, "step": 8860}, {"loss": 1.0451, "grad_norm": 1.3300775289535522, "learning_rate": 0.0002, "epoch": 5.7973856209150325, "step": 8870}, {"loss": 1.0989, "grad_norm": 1.612707257270813, "learning_rate": 0.0002, "epoch": 5.803921568627451, "step": 8880}, {"loss": 0.9119, "grad_norm": 0.9047797322273254, "learning_rate": 0.0002, "epoch": 5.810457516339869, "step": 8890}, {"loss": 0.989, "grad_norm": 1.0958741903305054, "learning_rate": 0.0002, "epoch": 5.816993464052287, "step": 8900}, {"loss": 1.1922, "grad_norm": 1.0099612474441528, "learning_rate": 0.0002, "epoch": 5.823529411764706, "step": 8910}, {"loss": 1.0623, "grad_norm": 0.8442328572273254, "learning_rate": 0.0002, "epoch": 5.830065359477124, "step": 8920}, {"loss": 0.9134, "grad_norm": 1.1388301849365234, "learning_rate": 0.0002, "epoch": 5.836601307189542, "step": 8930}, {"loss": 1.0019, "grad_norm": 0.8296026587486267, "learning_rate": 0.0002, "epoch": 5.8431372549019605, "step": 8940}, {"loss": 1.0363, "grad_norm": 1.0843533277511597, "learning_rate": 0.0002, "epoch": 5.849673202614379, "step": 8950}, {"loss": 1.0009, "grad_norm": 0.8496834635734558, "learning_rate": 0.0002, "epoch": 5.856209150326797, "step": 8960}, {"loss": 0.9927, "grad_norm": 1.6894690990447998, "learning_rate": 0.0002, "epoch": 5.862745098039216, "step": 8970}, {"loss": 1.0939, "grad_norm": 1.0012282133102417, "learning_rate": 0.0002, "epoch": 5.8692810457516345, "step": 8980}, {"loss": 0.9722, "grad_norm": 0.8521103262901306, "learning_rate": 0.0002, "epoch": 5.875816993464053, "step": 8990}, {"loss": 1.0885, "grad_norm": 1.246841311454773, "learning_rate": 0.0002, "epoch": 5.882352941176471, "step": 9000}, {"loss": 0.9702, "grad_norm": 0.9941892027854919, "learning_rate": 0.0002, "epoch": 5.888888888888889, "step": 9010}, {"loss": 0.8754, "grad_norm": 1.067413568496704, "learning_rate": 0.0002, "epoch": 5.895424836601308, "step": 9020}, {"loss": 1.0153, "grad_norm": 1.0045088529586792, "learning_rate": 0.0002, "epoch": 5.901960784313726, "step": 9030}, {"loss": 1.0134, "grad_norm": 1.383063554763794, "learning_rate": 0.0002, "epoch": 5.908496732026144, "step": 9040}, {"loss": 1.0845, "grad_norm": 0.8754428625106812, "learning_rate": 0.0002, "epoch": 5.915032679738562, "step": 9050}, {"loss": 0.9571, "grad_norm": 0.8577388525009155, "learning_rate": 0.0002, "epoch": 5.921568627450981, "step": 9060}, {"loss": 1.0532, "grad_norm": 0.8718975186347961, "learning_rate": 0.0002, "epoch": 5.928104575163399, "step": 9070}, {"loss": 1.0667, "grad_norm": 1.1762131452560425, "learning_rate": 0.0002, "epoch": 5.934640522875817, "step": 9080}, {"loss": 1.1114, "grad_norm": 1.1025866270065308, "learning_rate": 0.0002, "epoch": 5.9411764705882355, "step": 9090}, {"loss": 0.9155, "grad_norm": 1.0439870357513428, "learning_rate": 0.0002, "epoch": 5.947712418300654, "step": 9100}, {"loss": 1.0055, "grad_norm": 1.2411525249481201, "learning_rate": 0.0002, "epoch": 5.954248366013072, "step": 9110}, {"loss": 0.9747, "grad_norm": 1.0317714214324951, "learning_rate": 0.0002, "epoch": 5.96078431372549, "step": 9120}, {"loss": 1.0352, "grad_norm": 0.9880492091178894, "learning_rate": 0.0002, "epoch": 5.967320261437909, "step": 9130}, {"loss": 1.0459, "grad_norm": 0.9039815664291382, "learning_rate": 0.0002, "epoch": 5.973856209150327, "step": 9140}, {"loss": 1.0413, "grad_norm": 0.9049116373062134, "learning_rate": 0.0002, "epoch": 5.980392156862745, "step": 9150}, {"loss": 0.9792, "grad_norm": 0.996749222278595, "learning_rate": 0.0002, "epoch": 5.9869281045751634, "step": 9160}, {"loss": 0.8857, "grad_norm": 0.8716062307357788, "learning_rate": 0.0002, "epoch": 5.993464052287582, "step": 9170}, {"loss": 1.019, "grad_norm": 1.3081294298171997, "learning_rate": 0.0002, "epoch": 6.0, "step": 9180}]}
+{"epoch": 7.0, "step": 10710, "epoch_duration": 1677.8897953033447, "total_accumulated_duration": 11969.50663638115, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}, {"eval_loss": 1.4131312370300293, "eval_runtime": 31.8967, "eval_samples_per_second": 13.669, "eval_steps_per_second": 1.724, "epoch": 3.0, "step": 4590}, {"loss": 1.1283, "grad_norm": 0.6926420331001282, "learning_rate": 0.0002, "epoch": 3.0065359477124183, "step": 4600}, {"loss": 1.1537, "grad_norm": 0.6656355857849121, "learning_rate": 0.0002, "epoch": 3.0130718954248366, "step": 4610}, {"loss": 1.308, "grad_norm": 0.9901936650276184, "learning_rate": 0.0002, "epoch": 3.019607843137255, "step": 4620}, {"loss": 1.22, "grad_norm": 0.6713474988937378, "learning_rate": 0.0002, "epoch": 3.026143790849673, "step": 4630}, {"loss": 1.2249, "grad_norm": 0.6199324131011963, "learning_rate": 0.0002, "epoch": 3.0326797385620914, "step": 4640}, {"loss": 1.242, "grad_norm": 0.7180785536766052, "learning_rate": 0.0002, "epoch": 3.0392156862745097, "step": 4650}, {"loss": 1.1349, "grad_norm": 0.8256588578224182, "learning_rate": 0.0002, "epoch": 3.045751633986928, "step": 4660}, {"loss": 1.1431, "grad_norm": 0.6637389063835144, "learning_rate": 0.0002, "epoch": 3.052287581699346, "step": 4670}, {"loss": 1.1096, "grad_norm": 0.6980698108673096, "learning_rate": 0.0002, "epoch": 3.0588235294117645, "step": 4680}, {"loss": 1.196, "grad_norm": 0.8091534972190857, "learning_rate": 0.0002, "epoch": 3.065359477124183, "step": 4690}, {"loss": 1.1652, "grad_norm": 0.5715174078941345, "learning_rate": 0.0002, "epoch": 3.0718954248366015, "step": 4700}, {"loss": 1.1427, "grad_norm": 0.735639750957489, "learning_rate": 0.0002, "epoch": 3.0784313725490198, "step": 4710}, {"loss": 1.1522, "grad_norm": 0.7619708180427551, "learning_rate": 0.0002, "epoch": 3.084967320261438, "step": 4720}, {"loss": 1.0853, "grad_norm": 1.263566017150879, "learning_rate": 0.0002, "epoch": 3.0915032679738563, "step": 4730}, {"loss": 1.1348, "grad_norm": 0.6600871682167053, "learning_rate": 0.0002, "epoch": 3.0980392156862746, "step": 4740}, {"loss": 1.1766, "grad_norm": 0.717792809009552, "learning_rate": 0.0002, "epoch": 3.104575163398693, "step": 4750}, {"loss": 1.088, "grad_norm": 0.853714644908905, "learning_rate": 0.0002, "epoch": 3.111111111111111, "step": 4760}, {"loss": 1.2031, "grad_norm": 1.1004153490066528, "learning_rate": 0.0002, "epoch": 3.1176470588235294, "step": 4770}, {"loss": 1.3295, "grad_norm": 0.8566235899925232, "learning_rate": 0.0002, "epoch": 3.1241830065359477, "step": 4780}, {"loss": 1.2436, "grad_norm": 0.8315296173095703, "learning_rate": 0.0002, "epoch": 3.130718954248366, "step": 4790}, {"loss": 1.32, "grad_norm": 0.8020524978637695, "learning_rate": 0.0002, "epoch": 3.1372549019607843, "step": 4800}, {"loss": 1.1238, "grad_norm": 0.7564275860786438, "learning_rate": 0.0002, "epoch": 3.1437908496732025, "step": 4810}, {"loss": 1.1244, "grad_norm": 0.9077776670455933, "learning_rate": 0.0002, "epoch": 3.150326797385621, "step": 4820}, {"loss": 1.1399, "grad_norm": 0.6323099732398987, "learning_rate": 0.0002, "epoch": 3.156862745098039, "step": 4830}, {"loss": 1.1983, "grad_norm": 0.6625368595123291, "learning_rate": 0.0002, "epoch": 3.1633986928104574, "step": 4840}, {"loss": 1.066, "grad_norm": 0.8119261860847473, "learning_rate": 0.0002, "epoch": 3.1699346405228757, "step": 4850}, {"loss": 1.0224, "grad_norm": 0.6399450898170471, "learning_rate": 0.0002, "epoch": 3.176470588235294, "step": 4860}, {"loss": 1.2181, "grad_norm": 1.0659016370773315, "learning_rate": 0.0002, "epoch": 3.183006535947712, "step": 4870}, {"loss": 1.2914, "grad_norm": 0.8040369749069214, "learning_rate": 0.0002, "epoch": 3.189542483660131, "step": 4880}, {"loss": 1.1996, "grad_norm": 0.7784733176231384, "learning_rate": 0.0002, "epoch": 3.196078431372549, "step": 4890}, {"loss": 1.2051, "grad_norm": 0.9660294651985168, "learning_rate": 0.0002, "epoch": 3.2026143790849675, "step": 4900}, {"loss": 1.0419, "grad_norm": 1.0676977634429932, "learning_rate": 0.0002, "epoch": 3.2091503267973858, "step": 4910}, {"loss": 1.0083, "grad_norm": 0.5877565741539001, "learning_rate": 0.0002, "epoch": 3.215686274509804, "step": 4920}, {"loss": 1.1046, "grad_norm": 0.6164032816886902, "learning_rate": 0.0002, "epoch": 3.2222222222222223, "step": 4930}, {"loss": 1.1079, "grad_norm": 0.7627606987953186, "learning_rate": 0.0002, "epoch": 3.2287581699346406, "step": 4940}, {"loss": 1.2453, "grad_norm": 0.7442803978919983, "learning_rate": 0.0002, "epoch": 3.235294117647059, "step": 4950}, {"loss": 1.1087, "grad_norm": 0.7277812361717224, "learning_rate": 0.0002, "epoch": 3.241830065359477, "step": 4960}, {"loss": 1.2237, "grad_norm": 1.0301902294158936, "learning_rate": 0.0002, "epoch": 3.2483660130718954, "step": 4970}, {"loss": 1.1466, "grad_norm": 0.7798232436180115, "learning_rate": 0.0002, "epoch": 3.2549019607843137, "step": 4980}, {"loss": 1.2142, "grad_norm": 1.210265874862671, "learning_rate": 0.0002, "epoch": 3.261437908496732, "step": 4990}, {"loss": 1.1557, "grad_norm": 0.6677713990211487, "learning_rate": 0.0002, "epoch": 3.2679738562091503, "step": 5000}, {"loss": 1.3294, "grad_norm": 1.0524500608444214, "learning_rate": 0.0002, "epoch": 3.2745098039215685, "step": 5010}, {"loss": 1.1939, "grad_norm": 0.7091745734214783, "learning_rate": 0.0002, "epoch": 3.281045751633987, "step": 5020}, {"loss": 1.1891, "grad_norm": 0.8523224592208862, "learning_rate": 0.0002, "epoch": 3.287581699346405, "step": 5030}, {"loss": 1.1925, "grad_norm": 0.6120608448982239, "learning_rate": 0.0002, "epoch": 3.2941176470588234, "step": 5040}, {"loss": 1.0603, "grad_norm": 0.7437472939491272, "learning_rate": 0.0002, "epoch": 3.3006535947712417, "step": 5050}, {"loss": 1.1295, "grad_norm": 0.7611715197563171, "learning_rate": 0.0002, "epoch": 3.30718954248366, "step": 5060}, {"loss": 1.0531, "grad_norm": 0.7249704003334045, "learning_rate": 0.0002, "epoch": 3.313725490196078, "step": 5070}, {"loss": 1.2292, "grad_norm": 0.7316247820854187, "learning_rate": 0.0002, "epoch": 3.3202614379084965, "step": 5080}, {"loss": 1.1974, "grad_norm": 0.562412440776825, "learning_rate": 0.0002, "epoch": 3.326797385620915, "step": 5090}, {"loss": 1.0736, "grad_norm": 0.7052176594734192, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 5100}, {"loss": 1.122, "grad_norm": 0.7714211344718933, "learning_rate": 0.0002, "epoch": 3.3398692810457518, "step": 5110}, {"loss": 1.1684, "grad_norm": 1.0436055660247803, "learning_rate": 0.0002, "epoch": 3.34640522875817, "step": 5120}, {"loss": 1.0945, "grad_norm": 0.8867271542549133, "learning_rate": 0.0002, "epoch": 3.3529411764705883, "step": 5130}, {"loss": 1.159, "grad_norm": 0.8371267914772034, "learning_rate": 0.0002, "epoch": 3.3594771241830066, "step": 5140}, {"loss": 1.1073, "grad_norm": 0.7257837057113647, "learning_rate": 0.0002, "epoch": 3.366013071895425, "step": 5150}, {"loss": 1.1162, "grad_norm": 0.7102002501487732, "learning_rate": 0.0002, "epoch": 3.372549019607843, "step": 5160}, {"loss": 1.2056, "grad_norm": 0.7636350393295288, "learning_rate": 0.0002, "epoch": 3.3790849673202614, "step": 5170}, {"loss": 1.0708, "grad_norm": 0.6887359619140625, "learning_rate": 0.0002, "epoch": 3.3856209150326797, "step": 5180}, {"loss": 1.3807, "grad_norm": 0.8141424655914307, "learning_rate": 0.0002, "epoch": 3.392156862745098, "step": 5190}, {"loss": 1.1986, "grad_norm": 0.694423496723175, "learning_rate": 0.0002, "epoch": 3.3986928104575163, "step": 5200}, {"loss": 1.2945, "grad_norm": 0.914013683795929, "learning_rate": 0.0002, "epoch": 3.4052287581699345, "step": 5210}, {"loss": 1.1413, "grad_norm": 0.8503239750862122, "learning_rate": 0.0002, "epoch": 3.411764705882353, "step": 5220}, {"loss": 1.2696, "grad_norm": 0.6196836233139038, "learning_rate": 0.0002, "epoch": 3.418300653594771, "step": 5230}, {"loss": 1.2431, "grad_norm": 1.0760811567306519, "learning_rate": 0.0002, "epoch": 3.4248366013071894, "step": 5240}, {"loss": 1.1686, "grad_norm": 0.6524698138237, "learning_rate": 0.0002, "epoch": 3.431372549019608, "step": 5250}, {"loss": 1.2012, "grad_norm": 0.674467921257019, "learning_rate": 0.0002, "epoch": 3.4379084967320264, "step": 5260}, {"loss": 1.1015, "grad_norm": 0.7690372467041016, "learning_rate": 0.0002, "epoch": 3.4444444444444446, "step": 5270}, {"loss": 1.2511, "grad_norm": 0.8751813769340515, "learning_rate": 0.0002, "epoch": 3.450980392156863, "step": 5280}, {"loss": 1.1841, "grad_norm": 0.750407874584198, "learning_rate": 0.0002, "epoch": 3.457516339869281, "step": 5290}, {"loss": 1.0605, "grad_norm": 0.5991823077201843, "learning_rate": 0.0002, "epoch": 3.4640522875816995, "step": 5300}, {"loss": 1.2347, "grad_norm": 1.0164772272109985, "learning_rate": 0.0002, "epoch": 3.4705882352941178, "step": 5310}, {"loss": 1.2354, "grad_norm": 0.8704105019569397, "learning_rate": 0.0002, "epoch": 3.477124183006536, "step": 5320}, {"loss": 1.2169, "grad_norm": 0.709102213382721, "learning_rate": 0.0002, "epoch": 3.4836601307189543, "step": 5330}, {"loss": 1.2425, "grad_norm": 0.6273632049560547, "learning_rate": 0.0002, "epoch": 3.4901960784313726, "step": 5340}, {"loss": 1.1585, "grad_norm": 0.6807359457015991, "learning_rate": 0.0002, "epoch": 3.496732026143791, "step": 5350}, {"loss": 1.131, "grad_norm": 0.7085188627243042, "learning_rate": 0.0002, "epoch": 3.503267973856209, "step": 5360}, {"loss": 1.1159, "grad_norm": 0.6938307881355286, "learning_rate": 0.0002, "epoch": 3.5098039215686274, "step": 5370}, {"loss": 1.1397, "grad_norm": 0.8544146418571472, "learning_rate": 0.0002, "epoch": 3.5163398692810457, "step": 5380}, {"loss": 1.2181, "grad_norm": 0.7889642119407654, "learning_rate": 0.0002, "epoch": 3.522875816993464, "step": 5390}, {"loss": 1.1691, "grad_norm": 0.7858421206474304, "learning_rate": 0.0002, "epoch": 3.5294117647058822, "step": 5400}, {"loss": 1.2374, "grad_norm": 0.8547123074531555, "learning_rate": 0.0002, "epoch": 3.5359477124183005, "step": 5410}, {"loss": 1.196, "grad_norm": 0.8218181133270264, "learning_rate": 0.0002, "epoch": 3.542483660130719, "step": 5420}, {"loss": 1.1961, "grad_norm": 1.153623342514038, "learning_rate": 0.0002, "epoch": 3.549019607843137, "step": 5430}, {"loss": 1.156, "grad_norm": 1.1321099996566772, "learning_rate": 0.0002, "epoch": 3.5555555555555554, "step": 5440}, {"loss": 1.2224, "grad_norm": 0.9495334029197693, "learning_rate": 0.0002, "epoch": 3.5620915032679736, "step": 5450}, {"loss": 1.2869, "grad_norm": 0.8743821978569031, "learning_rate": 0.0002, "epoch": 3.568627450980392, "step": 5460}, {"loss": 1.1018, "grad_norm": 0.7513086795806885, "learning_rate": 0.0002, "epoch": 3.57516339869281, "step": 5470}, {"loss": 1.1082, "grad_norm": 1.0139480829238892, "learning_rate": 0.0002, "epoch": 3.581699346405229, "step": 5480}, {"loss": 1.1706, "grad_norm": 0.6615135073661804, "learning_rate": 0.0002, "epoch": 3.588235294117647, "step": 5490}, {"loss": 1.3906, "grad_norm": 1.180798888206482, "learning_rate": 0.0002, "epoch": 3.5947712418300655, "step": 5500}, {"loss": 1.2391, "grad_norm": 0.7085279226303101, "learning_rate": 0.0002, "epoch": 3.6013071895424837, "step": 5510}, {"loss": 1.1623, "grad_norm": 0.540268063545227, "learning_rate": 0.0002, "epoch": 3.607843137254902, "step": 5520}, {"loss": 1.2132, "grad_norm": 0.7905671000480652, "learning_rate": 0.0002, "epoch": 3.6143790849673203, "step": 5530}, {"loss": 1.2731, "grad_norm": 0.8457717299461365, "learning_rate": 0.0002, "epoch": 3.6209150326797386, "step": 5540}, {"loss": 1.1799, "grad_norm": 0.7102677822113037, "learning_rate": 0.0002, "epoch": 3.627450980392157, "step": 5550}, {"loss": 1.2394, "grad_norm": 0.7179514765739441, "learning_rate": 0.0002, "epoch": 3.633986928104575, "step": 5560}, {"loss": 1.2019, "grad_norm": 1.0854148864746094, "learning_rate": 0.0002, "epoch": 3.6405228758169934, "step": 5570}, {"loss": 1.1986, "grad_norm": 0.8209951519966125, "learning_rate": 0.0002, "epoch": 3.6470588235294117, "step": 5580}, {"loss": 1.2289, "grad_norm": 0.6944138407707214, "learning_rate": 0.0002, "epoch": 3.65359477124183, "step": 5590}, {"loss": 1.3226, "grad_norm": 0.7675473093986511, "learning_rate": 0.0002, "epoch": 3.6601307189542482, "step": 5600}, {"loss": 1.2866, "grad_norm": 0.6683364510536194, "learning_rate": 0.0002, "epoch": 3.6666666666666665, "step": 5610}, {"loss": 1.1099, "grad_norm": 0.7920727133750916, "learning_rate": 0.0002, "epoch": 3.6732026143790852, "step": 5620}, {"loss": 1.2287, "grad_norm": 0.9440218806266785, "learning_rate": 0.0002, "epoch": 3.6797385620915035, "step": 5630}, {"loss": 1.2444, "grad_norm": 0.6600824594497681, "learning_rate": 0.0002, "epoch": 3.686274509803922, "step": 5640}, {"loss": 1.191, "grad_norm": 0.6860619187355042, "learning_rate": 0.0002, "epoch": 3.69281045751634, "step": 5650}, {"loss": 1.1914, "grad_norm": 0.6579713225364685, "learning_rate": 0.0002, "epoch": 3.6993464052287583, "step": 5660}, {"loss": 1.1464, "grad_norm": 0.661081075668335, "learning_rate": 0.0002, "epoch": 3.7058823529411766, "step": 5670}, {"loss": 1.289, "grad_norm": 1.0968825817108154, "learning_rate": 0.0002, "epoch": 3.712418300653595, "step": 5680}, {"loss": 1.192, "grad_norm": 0.8066844940185547, "learning_rate": 0.0002, "epoch": 3.718954248366013, "step": 5690}, {"loss": 1.2322, "grad_norm": 0.8341682553291321, "learning_rate": 0.0002, "epoch": 3.7254901960784315, "step": 5700}, {"loss": 1.1473, "grad_norm": 0.6682852506637573, "learning_rate": 0.0002, "epoch": 3.7320261437908497, "step": 5710}, {"loss": 1.1566, "grad_norm": 0.898595929145813, "learning_rate": 0.0002, "epoch": 3.738562091503268, "step": 5720}, {"loss": 1.0919, "grad_norm": 0.6876054406166077, "learning_rate": 0.0002, "epoch": 3.7450980392156863, "step": 5730}, {"loss": 1.2302, "grad_norm": 0.7817103266716003, "learning_rate": 0.0002, "epoch": 3.7516339869281046, "step": 5740}, {"loss": 1.2439, "grad_norm": 0.5840168595314026, "learning_rate": 0.0002, "epoch": 3.758169934640523, "step": 5750}, {"loss": 1.1279, "grad_norm": 0.6263918876647949, "learning_rate": 0.0002, "epoch": 3.764705882352941, "step": 5760}, {"loss": 1.2023, "grad_norm": 0.7948952317237854, "learning_rate": 0.0002, "epoch": 3.7712418300653594, "step": 5770}, {"loss": 1.149, "grad_norm": 0.6700998544692993, "learning_rate": 0.0002, "epoch": 3.7777777777777777, "step": 5780}, {"loss": 1.3207, "grad_norm": 1.1169519424438477, "learning_rate": 0.0002, "epoch": 3.784313725490196, "step": 5790}, {"loss": 1.064, "grad_norm": 0.8354471325874329, "learning_rate": 0.0002, "epoch": 3.7908496732026142, "step": 5800}, {"loss": 1.2104, "grad_norm": 0.6304181814193726, "learning_rate": 0.0002, "epoch": 3.7973856209150325, "step": 5810}, {"loss": 1.2059, "grad_norm": 0.6919655799865723, "learning_rate": 0.0002, "epoch": 3.803921568627451, "step": 5820}, {"loss": 1.217, "grad_norm": 0.600385844707489, "learning_rate": 0.0002, "epoch": 3.810457516339869, "step": 5830}, {"loss": 1.2324, "grad_norm": 0.8406319618225098, "learning_rate": 0.0002, "epoch": 3.8169934640522873, "step": 5840}, {"loss": 1.2418, "grad_norm": 0.7594282031059265, "learning_rate": 0.0002, "epoch": 3.8235294117647056, "step": 5850}, {"loss": 1.1903, "grad_norm": 0.8179879784584045, "learning_rate": 0.0002, "epoch": 3.8300653594771243, "step": 5860}, {"loss": 1.255, "grad_norm": 1.141430377960205, "learning_rate": 0.0002, "epoch": 3.8366013071895426, "step": 5870}, {"loss": 1.1467, "grad_norm": 0.6595550775527954, "learning_rate": 0.0002, "epoch": 3.843137254901961, "step": 5880}, {"loss": 1.2378, "grad_norm": 0.7499435544013977, "learning_rate": 0.0002, "epoch": 3.849673202614379, "step": 5890}, {"loss": 1.217, "grad_norm": 0.7851517200469971, "learning_rate": 0.0002, "epoch": 3.8562091503267975, "step": 5900}, {"loss": 1.162, "grad_norm": 1.0533545017242432, "learning_rate": 0.0002, "epoch": 3.8627450980392157, "step": 5910}, {"loss": 1.3576, "grad_norm": 0.960086464881897, "learning_rate": 0.0002, "epoch": 3.869281045751634, "step": 5920}, {"loss": 1.151, "grad_norm": 0.9952049851417542, "learning_rate": 0.0002, "epoch": 3.8758169934640523, "step": 5930}, {"loss": 1.2027, "grad_norm": 0.7884191274642944, "learning_rate": 0.0002, "epoch": 3.8823529411764706, "step": 5940}, {"loss": 1.1796, "grad_norm": 0.7461766600608826, "learning_rate": 0.0002, "epoch": 3.888888888888889, "step": 5950}, {"loss": 1.2251, "grad_norm": 0.9594355821609497, "learning_rate": 0.0002, "epoch": 3.895424836601307, "step": 5960}, {"loss": 1.1164, "grad_norm": 0.8179471492767334, "learning_rate": 0.0002, "epoch": 3.9019607843137254, "step": 5970}, {"loss": 1.2421, "grad_norm": 0.8240267634391785, "learning_rate": 0.0002, "epoch": 3.9084967320261437, "step": 5980}, {"loss": 1.3076, "grad_norm": 0.7462618350982666, "learning_rate": 0.0002, "epoch": 3.915032679738562, "step": 5990}, {"loss": 1.2124, "grad_norm": 0.711207389831543, "learning_rate": 0.0002, "epoch": 3.9215686274509802, "step": 6000}, {"loss": 1.2119, "grad_norm": 0.6910956501960754, "learning_rate": 0.0002, "epoch": 3.928104575163399, "step": 6010}, {"loss": 1.2127, "grad_norm": 0.749093770980835, "learning_rate": 0.0002, "epoch": 3.9346405228758172, "step": 6020}, {"loss": 1.1542, "grad_norm": 1.3332762718200684, "learning_rate": 0.0002, "epoch": 3.9411764705882355, "step": 6030}, {"loss": 1.1442, "grad_norm": 0.71457439661026, "learning_rate": 0.0002, "epoch": 3.947712418300654, "step": 6040}, {"loss": 1.339, "grad_norm": 1.1205238103866577, "learning_rate": 0.0002, "epoch": 3.954248366013072, "step": 6050}, {"loss": 1.2962, "grad_norm": 0.6958928108215332, "learning_rate": 0.0002, "epoch": 3.9607843137254903, "step": 6060}, {"loss": 1.1802, "grad_norm": 0.7518056035041809, "learning_rate": 0.0002, "epoch": 3.9673202614379086, "step": 6070}, {"loss": 1.1179, "grad_norm": 0.8010755777359009, "learning_rate": 0.0002, "epoch": 3.973856209150327, "step": 6080}, {"loss": 1.2867, "grad_norm": 0.7492658495903015, "learning_rate": 0.0002, "epoch": 3.980392156862745, "step": 6090}, {"loss": 1.2113, "grad_norm": 0.900704562664032, "learning_rate": 0.0002, "epoch": 3.9869281045751634, "step": 6100}, {"loss": 1.1106, "grad_norm": 0.7997331619262695, "learning_rate": 0.0002, "epoch": 3.9934640522875817, "step": 6110}, {"loss": 1.1244, "grad_norm": 0.7163209319114685, "learning_rate": 0.0002, "epoch": 4.0, "step": 6120}, {"eval_loss": 1.4113320112228394, "eval_runtime": 33.7199, "eval_samples_per_second": 12.93, "eval_steps_per_second": 1.631, "epoch": 4.0, "step": 6120}, {"loss": 1.0423, "grad_norm": 0.9527022838592529, "learning_rate": 0.0002, "epoch": 4.006535947712418, "step": 6130}, {"loss": 1.101, "grad_norm": 0.7603210210800171, "learning_rate": 0.0002, "epoch": 4.0130718954248366, "step": 6140}, {"loss": 1.1834, "grad_norm": 1.127387523651123, "learning_rate": 0.0002, "epoch": 4.019607843137255, "step": 6150}, {"loss": 1.0734, "grad_norm": 0.8290133476257324, "learning_rate": 0.0002, "epoch": 4.026143790849673, "step": 6160}, {"loss": 1.0785, "grad_norm": 0.9912241101264954, "learning_rate": 0.0002, "epoch": 4.032679738562091, "step": 6170}, {"loss": 1.0719, "grad_norm": 0.947005033493042, "learning_rate": 0.0002, "epoch": 4.03921568627451, "step": 6180}, {"loss": 1.0835, "grad_norm": 0.707466185092926, "learning_rate": 0.0002, "epoch": 4.045751633986928, "step": 6190}, {"loss": 1.1079, "grad_norm": 1.0604327917099, "learning_rate": 0.0002, "epoch": 4.052287581699346, "step": 6200}, {"loss": 1.0375, "grad_norm": 0.7848685383796692, "learning_rate": 0.0002, "epoch": 4.0588235294117645, "step": 6210}, {"loss": 1.1167, "grad_norm": 0.8475256562232971, "learning_rate": 0.0002, "epoch": 4.065359477124183, "step": 6220}, {"loss": 1.1104, "grad_norm": 0.9759448766708374, "learning_rate": 0.0002, "epoch": 4.071895424836601, "step": 6230}, {"loss": 1.1538, "grad_norm": 0.9324519038200378, "learning_rate": 0.0002, "epoch": 4.078431372549019, "step": 6240}, {"loss": 1.0817, "grad_norm": 0.8723901510238647, "learning_rate": 0.0002, "epoch": 4.084967320261438, "step": 6250}, {"loss": 1.0977, "grad_norm": 0.8343415856361389, "learning_rate": 0.0002, "epoch": 4.091503267973856, "step": 6260}, {"loss": 0.9887, "grad_norm": 0.7490310072898865, "learning_rate": 0.0002, "epoch": 4.098039215686274, "step": 6270}, {"loss": 1.2084, "grad_norm": 0.8961182832717896, "learning_rate": 0.0002, "epoch": 4.104575163398692, "step": 6280}, {"loss": 1.1349, "grad_norm": 0.7124854922294617, "learning_rate": 0.0002, "epoch": 4.111111111111111, "step": 6290}, {"loss": 1.0081, "grad_norm": 0.8338138461112976, "learning_rate": 0.0002, "epoch": 4.117647058823529, "step": 6300}, {"loss": 1.1091, "grad_norm": 0.8075833320617676, "learning_rate": 0.0002, "epoch": 4.124183006535947, "step": 6310}, {"loss": 1.0193, "grad_norm": 0.8069391846656799, "learning_rate": 0.0002, "epoch": 4.130718954248366, "step": 6320}, {"loss": 0.948, "grad_norm": 0.9567893147468567, "learning_rate": 0.0002, "epoch": 4.137254901960785, "step": 6330}, {"loss": 1.0241, "grad_norm": 1.2184662818908691, "learning_rate": 0.0002, "epoch": 4.143790849673203, "step": 6340}, {"loss": 1.0756, "grad_norm": 1.030976414680481, "learning_rate": 0.0002, "epoch": 4.150326797385621, "step": 6350}, {"loss": 1.1124, "grad_norm": 0.9749957323074341, "learning_rate": 0.0002, "epoch": 4.1568627450980395, "step": 6360}, {"loss": 1.1038, "grad_norm": 0.7089483141899109, "learning_rate": 0.0002, "epoch": 4.163398692810458, "step": 6370}, {"loss": 1.2175, "grad_norm": 1.1084946393966675, "learning_rate": 0.0002, "epoch": 4.169934640522876, "step": 6380}, {"loss": 1.0274, "grad_norm": 0.7998497486114502, "learning_rate": 0.0002, "epoch": 4.176470588235294, "step": 6390}, {"loss": 1.005, "grad_norm": 0.8997811675071716, "learning_rate": 0.0002, "epoch": 4.183006535947713, "step": 6400}, {"loss": 1.0704, "grad_norm": 0.8359479904174805, "learning_rate": 0.0002, "epoch": 4.189542483660131, "step": 6410}, {"loss": 1.1056, "grad_norm": 0.9087472558021545, "learning_rate": 0.0002, "epoch": 4.196078431372549, "step": 6420}, {"loss": 1.0657, "grad_norm": 1.1100451946258545, "learning_rate": 0.0002, "epoch": 4.2026143790849675, "step": 6430}, {"loss": 1.1443, "grad_norm": 0.9376999735832214, "learning_rate": 0.0002, "epoch": 4.209150326797386, "step": 6440}, {"loss": 1.0862, "grad_norm": 0.8179266452789307, "learning_rate": 0.0002, "epoch": 4.215686274509804, "step": 6450}, {"loss": 1.0679, "grad_norm": 0.9953271746635437, "learning_rate": 0.0002, "epoch": 4.222222222222222, "step": 6460}, {"loss": 1.1034, "grad_norm": 0.8476650714874268, "learning_rate": 0.0002, "epoch": 4.228758169934641, "step": 6470}, {"loss": 1.2512, "grad_norm": 0.8406323194503784, "learning_rate": 0.0002, "epoch": 4.235294117647059, "step": 6480}, {"loss": 1.057, "grad_norm": 0.819134533405304, "learning_rate": 0.0002, "epoch": 4.241830065359477, "step": 6490}, {"loss": 1.1082, "grad_norm": 0.7764983773231506, "learning_rate": 0.0002, "epoch": 4.248366013071895, "step": 6500}, {"loss": 1.1593, "grad_norm": 0.8252112865447998, "learning_rate": 0.0002, "epoch": 4.254901960784314, "step": 6510}, {"loss": 1.1369, "grad_norm": 0.7941019535064697, "learning_rate": 0.0002, "epoch": 4.261437908496732, "step": 6520}, {"loss": 1.0296, "grad_norm": 0.7673905491828918, "learning_rate": 0.0002, "epoch": 4.26797385620915, "step": 6530}, {"loss": 1.1387, "grad_norm": 0.8749890327453613, "learning_rate": 0.0002, "epoch": 4.2745098039215685, "step": 6540}, {"loss": 1.0595, "grad_norm": 0.7343207597732544, "learning_rate": 0.0002, "epoch": 4.281045751633987, "step": 6550}, {"loss": 1.1715, "grad_norm": 1.2786651849746704, "learning_rate": 0.0002, "epoch": 4.287581699346405, "step": 6560}, {"loss": 1.0514, "grad_norm": 1.316875696182251, "learning_rate": 0.0002, "epoch": 4.294117647058823, "step": 6570}, {"loss": 1.1125, "grad_norm": 0.8349189162254333, "learning_rate": 0.0002, "epoch": 4.300653594771242, "step": 6580}, {"loss": 1.0732, "grad_norm": 0.7510647177696228, "learning_rate": 0.0002, "epoch": 4.30718954248366, "step": 6590}, {"loss": 1.1387, "grad_norm": 0.932420551776886, "learning_rate": 0.0002, "epoch": 4.313725490196078, "step": 6600}, {"loss": 1.1115, "grad_norm": 0.8510616421699524, "learning_rate": 0.0002, "epoch": 4.3202614379084965, "step": 6610}, {"loss": 1.0957, "grad_norm": 0.7661547064781189, "learning_rate": 0.0002, "epoch": 4.326797385620915, "step": 6620}, {"loss": 1.2064, "grad_norm": 1.0370930433273315, "learning_rate": 0.0002, "epoch": 4.333333333333333, "step": 6630}, {"loss": 1.1064, "grad_norm": 0.9302158951759338, "learning_rate": 0.0002, "epoch": 4.339869281045751, "step": 6640}, {"loss": 0.968, "grad_norm": 0.9203811883926392, "learning_rate": 0.0002, "epoch": 4.34640522875817, "step": 6650}, {"loss": 1.0123, "grad_norm": 0.9986332654953003, "learning_rate": 0.0002, "epoch": 4.352941176470588, "step": 6660}, {"loss": 1.1079, "grad_norm": 0.8001713156700134, "learning_rate": 0.0002, "epoch": 4.359477124183006, "step": 6670}, {"loss": 1.0248, "grad_norm": 0.829714298248291, "learning_rate": 0.0002, "epoch": 4.366013071895424, "step": 6680}, {"loss": 1.0389, "grad_norm": 0.8253079056739807, "learning_rate": 0.0002, "epoch": 4.372549019607844, "step": 6690}, {"loss": 1.1087, "grad_norm": 0.824666440486908, "learning_rate": 0.0002, "epoch": 4.379084967320262, "step": 6700}, {"loss": 1.1968, "grad_norm": 0.8872972130775452, "learning_rate": 0.0002, "epoch": 4.38562091503268, "step": 6710}, {"loss": 1.0474, "grad_norm": 0.8729761838912964, "learning_rate": 0.0002, "epoch": 4.392156862745098, "step": 6720}, {"loss": 1.0961, "grad_norm": 1.1367264986038208, "learning_rate": 0.0002, "epoch": 4.398692810457517, "step": 6730}, {"loss": 1.0184, "grad_norm": 0.9699058532714844, "learning_rate": 0.0002, "epoch": 4.405228758169935, "step": 6740}, {"loss": 1.006, "grad_norm": 0.8266763687133789, "learning_rate": 0.0002, "epoch": 4.411764705882353, "step": 6750}, {"loss": 1.0735, "grad_norm": 1.0249767303466797, "learning_rate": 0.0002, "epoch": 4.4183006535947715, "step": 6760}, {"loss": 1.1726, "grad_norm": 0.73606938123703, "learning_rate": 0.0002, "epoch": 4.42483660130719, "step": 6770}, {"loss": 1.1037, "grad_norm": 1.4050679206848145, "learning_rate": 0.0002, "epoch": 4.431372549019608, "step": 6780}, {"loss": 1.1418, "grad_norm": 1.1114081144332886, "learning_rate": 0.0002, "epoch": 4.437908496732026, "step": 6790}, {"loss": 0.9682, "grad_norm": 0.8031067848205566, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6800}, {"loss": 1.0753, "grad_norm": 0.8513566851615906, "learning_rate": 0.0002, "epoch": 4.450980392156863, "step": 6810}, {"loss": 1.1852, "grad_norm": 1.332741379737854, "learning_rate": 0.0002, "epoch": 4.457516339869281, "step": 6820}, {"loss": 1.0966, "grad_norm": 1.5032578706741333, "learning_rate": 0.0002, "epoch": 4.4640522875816995, "step": 6830}, {"loss": 1.1124, "grad_norm": 0.7677283883094788, "learning_rate": 0.0002, "epoch": 4.470588235294118, "step": 6840}, {"loss": 1.1501, "grad_norm": 0.989148736000061, "learning_rate": 0.0002, "epoch": 4.477124183006536, "step": 6850}, {"loss": 1.2239, "grad_norm": 1.5316275358200073, "learning_rate": 0.0002, "epoch": 4.483660130718954, "step": 6860}, {"loss": 1.1171, "grad_norm": 0.9427124261856079, "learning_rate": 0.0002, "epoch": 4.490196078431373, "step": 6870}, {"loss": 1.1314, "grad_norm": 1.215287685394287, "learning_rate": 0.0002, "epoch": 4.496732026143791, "step": 6880}, {"loss": 1.0809, "grad_norm": 0.7286760210990906, "learning_rate": 0.0002, "epoch": 4.503267973856209, "step": 6890}, {"loss": 1.0179, "grad_norm": 0.874829888343811, "learning_rate": 0.0002, "epoch": 4.509803921568627, "step": 6900}, {"loss": 1.0233, "grad_norm": 0.8058359622955322, "learning_rate": 0.0002, "epoch": 4.516339869281046, "step": 6910}, {"loss": 1.0463, "grad_norm": 1.248195767402649, "learning_rate": 0.0002, "epoch": 4.522875816993464, "step": 6920}, {"loss": 1.0347, "grad_norm": 0.8033645749092102, "learning_rate": 0.0002, "epoch": 4.529411764705882, "step": 6930}, {"loss": 1.1068, "grad_norm": 1.7361950874328613, "learning_rate": 0.0002, "epoch": 4.5359477124183005, "step": 6940}, {"loss": 0.9856, "grad_norm": 0.8058095574378967, "learning_rate": 0.0002, "epoch": 4.542483660130719, "step": 6950}, {"loss": 1.0057, "grad_norm": 1.254089593887329, "learning_rate": 0.0002, "epoch": 4.549019607843137, "step": 6960}, {"loss": 1.1723, "grad_norm": 0.9180455803871155, "learning_rate": 0.0002, "epoch": 4.555555555555555, "step": 6970}, {"loss": 1.0559, "grad_norm": 0.6677682399749756, "learning_rate": 0.0002, "epoch": 4.562091503267974, "step": 6980}, {"loss": 1.0453, "grad_norm": 0.8127354383468628, "learning_rate": 0.0002, "epoch": 4.568627450980392, "step": 6990}, {"loss": 1.0828, "grad_norm": 1.0263001918792725, "learning_rate": 0.0002, "epoch": 4.57516339869281, "step": 7000}, {"loss": 1.0703, "grad_norm": 0.9641909003257751, "learning_rate": 0.0002, "epoch": 4.5816993464052285, "step": 7010}, {"loss": 1.179, "grad_norm": 0.9440861344337463, "learning_rate": 0.0002, "epoch": 4.588235294117647, "step": 7020}, {"loss": 1.0931, "grad_norm": 0.9539011716842651, "learning_rate": 0.0002, "epoch": 4.594771241830065, "step": 7030}, {"loss": 1.0963, "grad_norm": 1.0449910163879395, "learning_rate": 0.0002, "epoch": 4.601307189542483, "step": 7040}, {"loss": 0.9944, "grad_norm": 0.8766893744468689, "learning_rate": 0.0002, "epoch": 4.607843137254902, "step": 7050}, {"loss": 1.0169, "grad_norm": 0.6983462572097778, "learning_rate": 0.0002, "epoch": 4.61437908496732, "step": 7060}, {"loss": 1.1778, "grad_norm": 0.9505505561828613, "learning_rate": 0.0002, "epoch": 4.620915032679738, "step": 7070}, {"loss": 1.121, "grad_norm": 1.2506657838821411, "learning_rate": 0.0002, "epoch": 4.627450980392156, "step": 7080}, {"loss": 1.1329, "grad_norm": 0.9602801203727722, "learning_rate": 0.0002, "epoch": 4.633986928104575, "step": 7090}, {"loss": 1.1499, "grad_norm": 0.7398977875709534, "learning_rate": 0.0002, "epoch": 4.640522875816993, "step": 7100}, {"loss": 1.0769, "grad_norm": 1.3862425088882446, "learning_rate": 0.0002, "epoch": 4.647058823529412, "step": 7110}, {"loss": 1.0571, "grad_norm": 1.1451990604400635, "learning_rate": 0.0002, "epoch": 4.65359477124183, "step": 7120}, {"loss": 1.1271, "grad_norm": 0.9010422229766846, "learning_rate": 0.0002, "epoch": 4.660130718954249, "step": 7130}, {"loss": 1.0165, "grad_norm": 0.7102518081665039, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 7140}, {"loss": 1.0819, "grad_norm": 0.7963796257972717, "learning_rate": 0.0002, "epoch": 4.673202614379085, "step": 7150}, {"loss": 1.1114, "grad_norm": 0.7726007699966431, "learning_rate": 0.0002, "epoch": 4.6797385620915035, "step": 7160}, {"loss": 1.2088, "grad_norm": 0.8097564578056335, "learning_rate": 0.0002, "epoch": 4.686274509803922, "step": 7170}, {"loss": 1.1386, "grad_norm": 0.9070925116539001, "learning_rate": 0.0002, "epoch": 4.69281045751634, "step": 7180}, {"loss": 1.0315, "grad_norm": 0.7543528079986572, "learning_rate": 0.0002, "epoch": 4.699346405228758, "step": 7190}, {"loss": 1.0984, "grad_norm": 0.9900904893875122, "learning_rate": 0.0002, "epoch": 4.705882352941177, "step": 7200}, {"loss": 1.1552, "grad_norm": 0.8033412098884583, "learning_rate": 0.0002, "epoch": 4.712418300653595, "step": 7210}, {"loss": 1.1773, "grad_norm": 0.8440839052200317, "learning_rate": 0.0002, "epoch": 4.718954248366013, "step": 7220}, {"loss": 1.1258, "grad_norm": 0.9325555562973022, "learning_rate": 0.0002, "epoch": 4.7254901960784315, "step": 7230}, {"loss": 1.1384, "grad_norm": 0.7881146669387817, "learning_rate": 0.0002, "epoch": 4.73202614379085, "step": 7240}, {"loss": 1.1219, "grad_norm": 0.884453296661377, "learning_rate": 0.0002, "epoch": 4.738562091503268, "step": 7250}, {"loss": 1.1036, "grad_norm": 0.9274539351463318, "learning_rate": 0.0002, "epoch": 4.745098039215686, "step": 7260}, {"loss": 1.0906, "grad_norm": 1.2367479801177979, "learning_rate": 0.0002, "epoch": 4.751633986928105, "step": 7270}, {"loss": 1.0741, "grad_norm": 0.9499821066856384, "learning_rate": 0.0002, "epoch": 4.758169934640523, "step": 7280}, {"loss": 1.1625, "grad_norm": 2.1918580532073975, "learning_rate": 0.0002, "epoch": 4.764705882352941, "step": 7290}, {"loss": 0.954, "grad_norm": 0.8221880793571472, "learning_rate": 0.0002, "epoch": 4.771241830065359, "step": 7300}, {"loss": 1.1358, "grad_norm": 0.871972918510437, "learning_rate": 0.0002, "epoch": 4.777777777777778, "step": 7310}, {"loss": 1.0599, "grad_norm": 0.8034510612487793, "learning_rate": 0.0002, "epoch": 4.784313725490196, "step": 7320}, {"loss": 1.1059, "grad_norm": 0.8959605693817139, "learning_rate": 0.0002, "epoch": 4.790849673202614, "step": 7330}, {"loss": 1.0176, "grad_norm": 1.2326215505599976, "learning_rate": 0.0002, "epoch": 4.7973856209150325, "step": 7340}, {"loss": 1.1095, "grad_norm": 0.9725791811943054, "learning_rate": 0.0002, "epoch": 4.803921568627451, "step": 7350}, {"loss": 1.1229, "grad_norm": 0.7240816354751587, "learning_rate": 0.0002, "epoch": 4.810457516339869, "step": 7360}, {"loss": 1.0669, "grad_norm": 0.8265769481658936, "learning_rate": 0.0002, "epoch": 4.816993464052287, "step": 7370}, {"loss": 1.042, "grad_norm": 0.8888696432113647, "learning_rate": 0.0002, "epoch": 4.823529411764706, "step": 7380}, {"loss": 1.0981, "grad_norm": 0.7776556015014648, "learning_rate": 0.0002, "epoch": 4.830065359477124, "step": 7390}, {"loss": 1.0819, "grad_norm": 0.8772371411323547, "learning_rate": 0.0002, "epoch": 4.836601307189542, "step": 7400}, {"loss": 1.0819, "grad_norm": 0.9786531925201416, "learning_rate": 0.0002, "epoch": 4.8431372549019605, "step": 7410}, {"loss": 1.1358, "grad_norm": 0.9059745073318481, "learning_rate": 0.0002, "epoch": 4.849673202614379, "step": 7420}, {"loss": 1.0324, "grad_norm": 0.7422552108764648, "learning_rate": 0.0002, "epoch": 4.856209150326797, "step": 7430}, {"loss": 1.0423, "grad_norm": 1.3040380477905273, "learning_rate": 0.0002, "epoch": 4.862745098039216, "step": 7440}, {"loss": 1.1161, "grad_norm": 1.3278473615646362, "learning_rate": 0.0002, "epoch": 4.8692810457516345, "step": 7450}, {"loss": 1.0713, "grad_norm": 1.2705849409103394, "learning_rate": 0.0002, "epoch": 4.875816993464053, "step": 7460}, {"loss": 1.0034, "grad_norm": 0.8837892413139343, "learning_rate": 0.0002, "epoch": 4.882352941176471, "step": 7470}, {"loss": 1.1716, "grad_norm": 0.8670691251754761, "learning_rate": 0.0002, "epoch": 4.888888888888889, "step": 7480}, {"loss": 1.1723, "grad_norm": 0.9662758111953735, "learning_rate": 0.0002, "epoch": 4.895424836601308, "step": 7490}, {"loss": 1.1056, "grad_norm": 0.8188302516937256, "learning_rate": 0.0002, "epoch": 4.901960784313726, "step": 7500}, {"loss": 1.0419, "grad_norm": 0.769442617893219, "learning_rate": 0.0002, "epoch": 4.908496732026144, "step": 7510}, {"loss": 1.1671, "grad_norm": 1.1465084552764893, "learning_rate": 0.0002, "epoch": 4.915032679738562, "step": 7520}, {"loss": 1.0768, "grad_norm": 1.253214955329895, "learning_rate": 0.0002, "epoch": 4.921568627450981, "step": 7530}, {"loss": 1.011, "grad_norm": 0.7922375202178955, "learning_rate": 0.0002, "epoch": 4.928104575163399, "step": 7540}, {"loss": 1.1256, "grad_norm": 0.8306851387023926, "learning_rate": 0.0002, "epoch": 4.934640522875817, "step": 7550}, {"loss": 1.206, "grad_norm": 0.8486151099205017, "learning_rate": 0.0002, "epoch": 4.9411764705882355, "step": 7560}, {"loss": 1.0161, "grad_norm": 1.2601467370986938, "learning_rate": 0.0002, "epoch": 4.947712418300654, "step": 7570}, {"loss": 1.1078, "grad_norm": 0.7980747818946838, "learning_rate": 0.0002, "epoch": 4.954248366013072, "step": 7580}, {"loss": 1.0607, "grad_norm": 0.8653254508972168, "learning_rate": 0.0002, "epoch": 4.96078431372549, "step": 7590}, {"loss": 1.0292, "grad_norm": 0.9680571556091309, "learning_rate": 0.0002, "epoch": 4.967320261437909, "step": 7600}, {"loss": 1.1795, "grad_norm": 0.9554466605186462, "learning_rate": 0.0002, "epoch": 4.973856209150327, "step": 7610}, {"loss": 1.0935, "grad_norm": 1.3693897724151611, "learning_rate": 0.0002, "epoch": 4.980392156862745, "step": 7620}, {"loss": 1.0838, "grad_norm": 0.7809282541275024, "learning_rate": 0.0002, "epoch": 4.9869281045751634, "step": 7630}, {"loss": 1.0844, "grad_norm": 0.7528006434440613, "learning_rate": 0.0002, "epoch": 4.993464052287582, "step": 7640}, {"loss": 0.9951, "grad_norm": 1.7491309642791748, "learning_rate": 0.0002, "epoch": 5.0, "step": 7650}, {"eval_loss": 1.4197258949279785, "eval_runtime": 33.6327, "eval_samples_per_second": 12.964, "eval_steps_per_second": 1.635, "epoch": 5.0, "step": 7650}, {"loss": 0.9744, "grad_norm": 0.8840063214302063, "learning_rate": 0.0002, "epoch": 5.006535947712418, "step": 7660}, {"loss": 1.0274, "grad_norm": 1.0118401050567627, "learning_rate": 0.0002, "epoch": 5.0130718954248366, "step": 7670}, {"loss": 1.1667, "grad_norm": 1.0040518045425415, "learning_rate": 0.0002, "epoch": 5.019607843137255, "step": 7680}, {"loss": 0.9426, "grad_norm": 0.7541199922561646, "learning_rate": 0.0002, "epoch": 5.026143790849673, "step": 7690}, {"loss": 1.0797, "grad_norm": 0.9106482863426208, "learning_rate": 0.0002, "epoch": 5.032679738562091, "step": 7700}, {"loss": 1.0096, "grad_norm": 1.3691469430923462, "learning_rate": 0.0002, "epoch": 5.03921568627451, "step": 7710}, {"loss": 0.9889, "grad_norm": 0.9449689388275146, "learning_rate": 0.0002, "epoch": 5.045751633986928, "step": 7720}, {"loss": 0.9087, "grad_norm": 1.1678508520126343, "learning_rate": 0.0002, "epoch": 5.052287581699346, "step": 7730}, {"loss": 1.0556, "grad_norm": 1.1296145915985107, "learning_rate": 0.0002, "epoch": 5.0588235294117645, "step": 7740}, {"loss": 0.9339, "grad_norm": 0.7863904237747192, "learning_rate": 0.0002, "epoch": 5.065359477124183, "step": 7750}, {"loss": 1.0135, "grad_norm": 0.8691433072090149, "learning_rate": 0.0002, "epoch": 5.071895424836601, "step": 7760}, {"loss": 0.9776, "grad_norm": 1.0722088813781738, "learning_rate": 0.0002, "epoch": 5.078431372549019, "step": 7770}, {"loss": 1.0595, "grad_norm": 0.9625038504600525, "learning_rate": 0.0002, "epoch": 5.084967320261438, "step": 7780}, {"loss": 1.0241, "grad_norm": 1.2618783712387085, "learning_rate": 0.0002, "epoch": 5.091503267973856, "step": 7790}, {"loss": 0.9396, "grad_norm": 0.9970650672912598, "learning_rate": 0.0002, "epoch": 5.098039215686274, "step": 7800}, {"loss": 0.9186, "grad_norm": 1.3946677446365356, "learning_rate": 0.0002, "epoch": 5.104575163398692, "step": 7810}, {"loss": 0.9957, "grad_norm": 1.0260052680969238, "learning_rate": 0.0002, "epoch": 5.111111111111111, "step": 7820}, {"loss": 0.9865, "grad_norm": 1.105521559715271, "learning_rate": 0.0002, "epoch": 5.117647058823529, "step": 7830}, {"loss": 0.9788, "grad_norm": 1.003641128540039, "learning_rate": 0.0002, "epoch": 5.124183006535947, "step": 7840}, {"loss": 0.9688, "grad_norm": 1.0315021276474, "learning_rate": 0.0002, "epoch": 5.130718954248366, "step": 7850}, {"loss": 1.0001, "grad_norm": 0.9469530582427979, "learning_rate": 0.0002, "epoch": 5.137254901960785, "step": 7860}, {"loss": 0.9659, "grad_norm": 1.3244667053222656, "learning_rate": 0.0002, "epoch": 5.143790849673203, "step": 7870}, {"loss": 0.9657, "grad_norm": 1.1732033491134644, "learning_rate": 0.0002, "epoch": 5.150326797385621, "step": 7880}, {"loss": 0.9978, "grad_norm": 1.3129149675369263, "learning_rate": 0.0002, "epoch": 5.1568627450980395, "step": 7890}, {"loss": 0.9894, "grad_norm": 0.8589454293251038, "learning_rate": 0.0002, "epoch": 5.163398692810458, "step": 7900}, {"loss": 1.0161, "grad_norm": 0.8954233527183533, "learning_rate": 0.0002, "epoch": 5.169934640522876, "step": 7910}, {"loss": 0.8741, "grad_norm": 0.7426522970199585, "learning_rate": 0.0002, "epoch": 5.176470588235294, "step": 7920}, {"loss": 1.0106, "grad_norm": 1.1990121603012085, "learning_rate": 0.0002, "epoch": 5.183006535947713, "step": 7930}, {"loss": 0.9453, "grad_norm": 0.8867580890655518, "learning_rate": 0.0002, "epoch": 5.189542483660131, "step": 7940}, {"loss": 0.9727, "grad_norm": 1.016276478767395, "learning_rate": 0.0002, "epoch": 5.196078431372549, "step": 7950}, {"loss": 0.9908, "grad_norm": 1.0210685729980469, "learning_rate": 0.0002, "epoch": 5.2026143790849675, "step": 7960}, {"loss": 1.0522, "grad_norm": 1.0093122720718384, "learning_rate": 0.0002, "epoch": 5.209150326797386, "step": 7970}, {"loss": 1.0055, "grad_norm": 0.9746801853179932, "learning_rate": 0.0002, "epoch": 5.215686274509804, "step": 7980}, {"loss": 1.0611, "grad_norm": 0.9113537073135376, "learning_rate": 0.0002, "epoch": 5.222222222222222, "step": 7990}, {"loss": 0.9167, "grad_norm": 1.2782206535339355, "learning_rate": 0.0002, "epoch": 5.228758169934641, "step": 8000}, {"loss": 1.0212, "grad_norm": 1.3223118782043457, "learning_rate": 0.0002, "epoch": 5.235294117647059, "step": 8010}, {"loss": 0.9244, "grad_norm": 0.7898629307746887, "learning_rate": 0.0002, "epoch": 5.241830065359477, "step": 8020}, {"loss": 1.0574, "grad_norm": 0.9822350740432739, "learning_rate": 0.0002, "epoch": 5.248366013071895, "step": 8030}, {"loss": 1.0102, "grad_norm": 1.5114340782165527, "learning_rate": 0.0002, "epoch": 5.254901960784314, "step": 8040}, {"loss": 0.9816, "grad_norm": 0.859006941318512, "learning_rate": 0.0002, "epoch": 5.261437908496732, "step": 8050}, {"loss": 0.9445, "grad_norm": 1.0495043992996216, "learning_rate": 0.0002, "epoch": 5.26797385620915, "step": 8060}, {"loss": 0.9724, "grad_norm": 1.329483151435852, "learning_rate": 0.0002, "epoch": 5.2745098039215685, "step": 8070}, {"loss": 0.9296, "grad_norm": 1.1333061456680298, "learning_rate": 0.0002, "epoch": 5.281045751633987, "step": 8080}, {"loss": 0.9577, "grad_norm": 0.8153108358383179, "learning_rate": 0.0002, "epoch": 5.287581699346405, "step": 8090}, {"loss": 0.9002, "grad_norm": 0.9395004510879517, "learning_rate": 0.0002, "epoch": 5.294117647058823, "step": 8100}, {"loss": 1.0371, "grad_norm": 0.8907593488693237, "learning_rate": 0.0002, "epoch": 5.300653594771242, "step": 8110}, {"loss": 0.9301, "grad_norm": 0.9808667898178101, "learning_rate": 0.0002, "epoch": 5.30718954248366, "step": 8120}, {"loss": 1.0136, "grad_norm": 0.984779417514801, "learning_rate": 0.0002, "epoch": 5.313725490196078, "step": 8130}, {"loss": 0.9621, "grad_norm": 0.9787270426750183, "learning_rate": 0.0002, "epoch": 5.3202614379084965, "step": 8140}, {"loss": 0.9336, "grad_norm": 0.9857710599899292, "learning_rate": 0.0002, "epoch": 5.326797385620915, "step": 8150}, {"loss": 0.9884, "grad_norm": 0.9774303436279297, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 8160}, {"loss": 1.0561, "grad_norm": 0.677925169467926, "learning_rate": 0.0002, "epoch": 5.339869281045751, "step": 8170}, {"loss": 1.1345, "grad_norm": 0.9576456546783447, "learning_rate": 0.0002, "epoch": 5.34640522875817, "step": 8180}, {"loss": 0.9554, "grad_norm": 1.8970937728881836, "learning_rate": 0.0002, "epoch": 5.352941176470588, "step": 8190}, {"loss": 1.0474, "grad_norm": 0.9458389282226562, "learning_rate": 0.0002, "epoch": 5.359477124183006, "step": 8200}, {"loss": 1.0365, "grad_norm": 1.761794924736023, "learning_rate": 0.0002, "epoch": 5.366013071895424, "step": 8210}, {"loss": 0.9426, "grad_norm": 1.0693724155426025, "learning_rate": 0.0002, "epoch": 5.372549019607844, "step": 8220}, {"loss": 1.0299, "grad_norm": 0.9025877714157104, "learning_rate": 0.0002, "epoch": 5.379084967320262, "step": 8230}, {"loss": 0.9652, "grad_norm": 1.258857250213623, "learning_rate": 0.0002, "epoch": 5.38562091503268, "step": 8240}, {"loss": 0.9735, "grad_norm": 1.084849238395691, "learning_rate": 0.0002, "epoch": 5.392156862745098, "step": 8250}, {"loss": 0.9999, "grad_norm": 0.9530340433120728, "learning_rate": 0.0002, "epoch": 5.398692810457517, "step": 8260}, {"loss": 1.0268, "grad_norm": 0.830240786075592, "learning_rate": 0.0002, "epoch": 5.405228758169935, "step": 8270}, {"loss": 1.0332, "grad_norm": 1.5807015895843506, "learning_rate": 0.0002, "epoch": 5.411764705882353, "step": 8280}, {"loss": 0.9146, "grad_norm": 0.9486905336380005, "learning_rate": 0.0002, "epoch": 5.4183006535947715, "step": 8290}, {"loss": 1.0336, "grad_norm": 1.0415093898773193, "learning_rate": 0.0002, "epoch": 5.42483660130719, "step": 8300}, {"loss": 0.8933, "grad_norm": 1.0501102209091187, "learning_rate": 0.0002, "epoch": 5.431372549019608, "step": 8310}, {"loss": 0.9983, "grad_norm": 0.9751836061477661, "learning_rate": 0.0002, "epoch": 5.437908496732026, "step": 8320}, {"loss": 1.0755, "grad_norm": 1.5529173612594604, "learning_rate": 0.0002, "epoch": 5.444444444444445, "step": 8330}, {"loss": 0.9814, "grad_norm": 0.8314350247383118, "learning_rate": 0.0002, "epoch": 5.450980392156863, "step": 8340}, {"loss": 1.0596, "grad_norm": 1.2555103302001953, "learning_rate": 0.0002, "epoch": 5.457516339869281, "step": 8350}, {"loss": 1.0127, "grad_norm": 0.9408367872238159, "learning_rate": 0.0002, "epoch": 5.4640522875816995, "step": 8360}, {"loss": 0.9241, "grad_norm": 0.9483312964439392, "learning_rate": 0.0002, "epoch": 5.470588235294118, "step": 8370}, {"loss": 0.9678, "grad_norm": 0.957905650138855, "learning_rate": 0.0002, "epoch": 5.477124183006536, "step": 8380}, {"loss": 1.0985, "grad_norm": 1.4000147581100464, "learning_rate": 0.0002, "epoch": 5.483660130718954, "step": 8390}, {"loss": 0.9966, "grad_norm": 1.7032461166381836, "learning_rate": 0.0002, "epoch": 5.490196078431373, "step": 8400}, {"loss": 0.9539, "grad_norm": 0.8978716731071472, "learning_rate": 0.0002, "epoch": 5.496732026143791, "step": 8410}, {"loss": 0.9544, "grad_norm": 0.8659300804138184, "learning_rate": 0.0002, "epoch": 5.503267973856209, "step": 8420}, {"loss": 1.0526, "grad_norm": 1.3629727363586426, "learning_rate": 0.0002, "epoch": 5.509803921568627, "step": 8430}, {"loss": 0.9696, "grad_norm": 1.2741984128952026, "learning_rate": 0.0002, "epoch": 5.516339869281046, "step": 8440}, {"loss": 1.0191, "grad_norm": 1.3867180347442627, "learning_rate": 0.0002, "epoch": 5.522875816993464, "step": 8450}, {"loss": 1.0835, "grad_norm": 1.0662001371383667, "learning_rate": 0.0002, "epoch": 5.529411764705882, "step": 8460}, {"loss": 0.9779, "grad_norm": 1.7005380392074585, "learning_rate": 0.0002, "epoch": 5.5359477124183005, "step": 8470}, {"loss": 1.0221, "grad_norm": 1.3730385303497314, "learning_rate": 0.0002, "epoch": 5.542483660130719, "step": 8480}, {"loss": 0.9586, "grad_norm": 1.7737441062927246, "learning_rate": 0.0002, "epoch": 5.549019607843137, "step": 8490}, {"loss": 0.9729, "grad_norm": 0.907487690448761, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 8500}, {"loss": 0.9891, "grad_norm": 0.8882441520690918, "learning_rate": 0.0002, "epoch": 5.562091503267974, "step": 8510}, {"loss": 0.973, "grad_norm": 0.8655388951301575, "learning_rate": 0.0002, "epoch": 5.568627450980392, "step": 8520}, {"loss": 0.9523, "grad_norm": 1.379992961883545, "learning_rate": 0.0002, "epoch": 5.57516339869281, "step": 8530}, {"loss": 1.0174, "grad_norm": 1.0021201372146606, "learning_rate": 0.0002, "epoch": 5.5816993464052285, "step": 8540}, {"loss": 1.0113, "grad_norm": 1.2636926174163818, "learning_rate": 0.0002, "epoch": 5.588235294117647, "step": 8550}, {"loss": 1.0243, "grad_norm": 1.279025912284851, "learning_rate": 0.0002, "epoch": 5.594771241830065, "step": 8560}, {"loss": 0.9917, "grad_norm": 0.8885834217071533, "learning_rate": 0.0002, "epoch": 5.601307189542483, "step": 8570}, {"loss": 0.9849, "grad_norm": 1.1975032091140747, "learning_rate": 0.0002, "epoch": 5.607843137254902, "step": 8580}, {"loss": 1.0363, "grad_norm": 1.005470871925354, "learning_rate": 0.0002, "epoch": 5.61437908496732, "step": 8590}, {"loss": 0.9947, "grad_norm": 1.104286551475525, "learning_rate": 0.0002, "epoch": 5.620915032679738, "step": 8600}, {"loss": 1.0585, "grad_norm": 1.435445785522461, "learning_rate": 0.0002, "epoch": 5.627450980392156, "step": 8610}, {"loss": 0.9156, "grad_norm": 1.0270172357559204, "learning_rate": 0.0002, "epoch": 5.633986928104575, "step": 8620}, {"loss": 1.0522, "grad_norm": 1.0929527282714844, "learning_rate": 0.0002, "epoch": 5.640522875816993, "step": 8630}, {"loss": 0.9694, "grad_norm": 1.1061221361160278, "learning_rate": 0.0002, "epoch": 5.647058823529412, "step": 8640}, {"loss": 1.0826, "grad_norm": 0.9563149213790894, "learning_rate": 0.0002, "epoch": 5.65359477124183, "step": 8650}, {"loss": 1.0042, "grad_norm": 1.0434954166412354, "learning_rate": 0.0002, "epoch": 5.660130718954249, "step": 8660}, {"loss": 0.9463, "grad_norm": 1.3695117235183716, "learning_rate": 0.0002, "epoch": 5.666666666666667, "step": 8670}, {"loss": 0.9441, "grad_norm": 1.0540564060211182, "learning_rate": 0.0002, "epoch": 5.673202614379085, "step": 8680}, {"loss": 0.9755, "grad_norm": 1.5942492485046387, "learning_rate": 0.0002, "epoch": 5.6797385620915035, "step": 8690}, {"loss": 1.0071, "grad_norm": 0.9485495090484619, "learning_rate": 0.0002, "epoch": 5.686274509803922, "step": 8700}, {"loss": 0.9998, "grad_norm": 1.1483162641525269, "learning_rate": 0.0002, "epoch": 5.69281045751634, "step": 8710}, {"loss": 0.9578, "grad_norm": 0.9075471758842468, "learning_rate": 0.0002, "epoch": 5.699346405228758, "step": 8720}, {"loss": 0.9488, "grad_norm": 1.7908551692962646, "learning_rate": 0.0002, "epoch": 5.705882352941177, "step": 8730}, {"loss": 1.0163, "grad_norm": 0.8867162466049194, "learning_rate": 0.0002, "epoch": 5.712418300653595, "step": 8740}, {"loss": 1.0041, "grad_norm": 1.7165148258209229, "learning_rate": 0.0002, "epoch": 5.718954248366013, "step": 8750}, {"loss": 1.1061, "grad_norm": 0.9529356956481934, "learning_rate": 0.0002, "epoch": 5.7254901960784315, "step": 8760}, {"loss": 1.1119, "grad_norm": 1.01852548122406, "learning_rate": 0.0002, "epoch": 5.73202614379085, "step": 8770}, {"loss": 1.0471, "grad_norm": 0.9538423418998718, "learning_rate": 0.0002, "epoch": 5.738562091503268, "step": 8780}, {"loss": 1.0913, "grad_norm": 0.9007737636566162, "learning_rate": 0.0002, "epoch": 5.745098039215686, "step": 8790}, {"loss": 0.9766, "grad_norm": 0.9107874035835266, "learning_rate": 0.0002, "epoch": 5.751633986928105, "step": 8800}, {"loss": 0.9212, "grad_norm": 0.7379238605499268, "learning_rate": 0.0002, "epoch": 5.758169934640523, "step": 8810}, {"loss": 1.0966, "grad_norm": 1.072645902633667, "learning_rate": 0.0002, "epoch": 5.764705882352941, "step": 8820}, {"loss": 1.0845, "grad_norm": 1.002008080482483, "learning_rate": 0.0002, "epoch": 5.771241830065359, "step": 8830}, {"loss": 0.9978, "grad_norm": 1.0435924530029297, "learning_rate": 0.0002, "epoch": 5.777777777777778, "step": 8840}, {"loss": 0.9458, "grad_norm": 0.9874551296234131, "learning_rate": 0.0002, "epoch": 5.784313725490196, "step": 8850}, {"loss": 1.1241, "grad_norm": 1.1729662418365479, "learning_rate": 0.0002, "epoch": 5.790849673202614, "step": 8860}, {"loss": 1.0451, "grad_norm": 1.3300775289535522, "learning_rate": 0.0002, "epoch": 5.7973856209150325, "step": 8870}, {"loss": 1.0989, "grad_norm": 1.612707257270813, "learning_rate": 0.0002, "epoch": 5.803921568627451, "step": 8880}, {"loss": 0.9119, "grad_norm": 0.9047797322273254, "learning_rate": 0.0002, "epoch": 5.810457516339869, "step": 8890}, {"loss": 0.989, "grad_norm": 1.0958741903305054, "learning_rate": 0.0002, "epoch": 5.816993464052287, "step": 8900}, {"loss": 1.1922, "grad_norm": 1.0099612474441528, "learning_rate": 0.0002, "epoch": 5.823529411764706, "step": 8910}, {"loss": 1.0623, "grad_norm": 0.8442328572273254, "learning_rate": 0.0002, "epoch": 5.830065359477124, "step": 8920}, {"loss": 0.9134, "grad_norm": 1.1388301849365234, "learning_rate": 0.0002, "epoch": 5.836601307189542, "step": 8930}, {"loss": 1.0019, "grad_norm": 0.8296026587486267, "learning_rate": 0.0002, "epoch": 5.8431372549019605, "step": 8940}, {"loss": 1.0363, "grad_norm": 1.0843533277511597, "learning_rate": 0.0002, "epoch": 5.849673202614379, "step": 8950}, {"loss": 1.0009, "grad_norm": 0.8496834635734558, "learning_rate": 0.0002, "epoch": 5.856209150326797, "step": 8960}, {"loss": 0.9927, "grad_norm": 1.6894690990447998, "learning_rate": 0.0002, "epoch": 5.862745098039216, "step": 8970}, {"loss": 1.0939, "grad_norm": 1.0012282133102417, "learning_rate": 0.0002, "epoch": 5.8692810457516345, "step": 8980}, {"loss": 0.9722, "grad_norm": 0.8521103262901306, "learning_rate": 0.0002, "epoch": 5.875816993464053, "step": 8990}, {"loss": 1.0885, "grad_norm": 1.246841311454773, "learning_rate": 0.0002, "epoch": 5.882352941176471, "step": 9000}, {"loss": 0.9702, "grad_norm": 0.9941892027854919, "learning_rate": 0.0002, "epoch": 5.888888888888889, "step": 9010}, {"loss": 0.8754, "grad_norm": 1.067413568496704, "learning_rate": 0.0002, "epoch": 5.895424836601308, "step": 9020}, {"loss": 1.0153, "grad_norm": 1.0045088529586792, "learning_rate": 0.0002, "epoch": 5.901960784313726, "step": 9030}, {"loss": 1.0134, "grad_norm": 1.383063554763794, "learning_rate": 0.0002, "epoch": 5.908496732026144, "step": 9040}, {"loss": 1.0845, "grad_norm": 0.8754428625106812, "learning_rate": 0.0002, "epoch": 5.915032679738562, "step": 9050}, {"loss": 0.9571, "grad_norm": 0.8577388525009155, "learning_rate": 0.0002, "epoch": 5.921568627450981, "step": 9060}, {"loss": 1.0532, "grad_norm": 0.8718975186347961, "learning_rate": 0.0002, "epoch": 5.928104575163399, "step": 9070}, {"loss": 1.0667, "grad_norm": 1.1762131452560425, "learning_rate": 0.0002, "epoch": 5.934640522875817, "step": 9080}, {"loss": 1.1114, "grad_norm": 1.1025866270065308, "learning_rate": 0.0002, "epoch": 5.9411764705882355, "step": 9090}, {"loss": 0.9155, "grad_norm": 1.0439870357513428, "learning_rate": 0.0002, "epoch": 5.947712418300654, "step": 9100}, {"loss": 1.0055, "grad_norm": 1.2411525249481201, "learning_rate": 0.0002, "epoch": 5.954248366013072, "step": 9110}, {"loss": 0.9747, "grad_norm": 1.0317714214324951, "learning_rate": 0.0002, "epoch": 5.96078431372549, "step": 9120}, {"loss": 1.0352, "grad_norm": 0.9880492091178894, "learning_rate": 0.0002, "epoch": 5.967320261437909, "step": 9130}, {"loss": 1.0459, "grad_norm": 0.9039815664291382, "learning_rate": 0.0002, "epoch": 5.973856209150327, "step": 9140}, {"loss": 1.0413, "grad_norm": 0.9049116373062134, "learning_rate": 0.0002, "epoch": 5.980392156862745, "step": 9150}, {"loss": 0.9792, "grad_norm": 0.996749222278595, "learning_rate": 0.0002, "epoch": 5.9869281045751634, "step": 9160}, {"loss": 0.8857, "grad_norm": 0.8716062307357788, "learning_rate": 0.0002, "epoch": 5.993464052287582, "step": 9170}, {"loss": 1.019, "grad_norm": 1.3081294298171997, "learning_rate": 0.0002, "epoch": 6.0, "step": 9180}, {"eval_loss": 1.45111083984375, "eval_runtime": 34.7121, "eval_samples_per_second": 12.56, "eval_steps_per_second": 1.584, "epoch": 6.0, "step": 9180}, {"loss": 0.9306, "grad_norm": 1.1378029584884644, "learning_rate": 0.0002, "epoch": 6.006535947712418, "step": 9190}, {"loss": 0.8794, "grad_norm": 1.2921233177185059, "learning_rate": 0.0002, "epoch": 6.0130718954248366, "step": 9200}, {"loss": 0.8145, "grad_norm": 1.039211630821228, "learning_rate": 0.0002, "epoch": 6.019607843137255, "step": 9210}, {"loss": 0.8524, "grad_norm": 0.9715196490287781, "learning_rate": 0.0002, "epoch": 6.026143790849673, "step": 9220}, {"loss": 1.035, "grad_norm": 1.220642328262329, "learning_rate": 0.0002, "epoch": 6.032679738562091, "step": 9230}, {"loss": 0.8468, "grad_norm": 0.854360044002533, "learning_rate": 0.0002, "epoch": 6.03921568627451, "step": 9240}, {"loss": 0.8534, "grad_norm": 0.8806933164596558, "learning_rate": 0.0002, "epoch": 6.045751633986928, "step": 9250}, {"loss": 0.8305, "grad_norm": 1.4315874576568604, "learning_rate": 0.0002, "epoch": 6.052287581699346, "step": 9260}, {"loss": 0.8462, "grad_norm": 0.9382007122039795, "learning_rate": 0.0002, "epoch": 6.0588235294117645, "step": 9270}, {"loss": 0.9653, "grad_norm": 1.2184561491012573, "learning_rate": 0.0002, "epoch": 6.065359477124183, "step": 9280}, {"loss": 0.8806, "grad_norm": 1.2331548929214478, "learning_rate": 0.0002, "epoch": 6.071895424836601, "step": 9290}, {"loss": 0.8354, "grad_norm": 1.1112796068191528, "learning_rate": 0.0002, "epoch": 6.078431372549019, "step": 9300}, {"loss": 0.8008, "grad_norm": 1.4753731489181519, "learning_rate": 0.0002, "epoch": 6.084967320261438, "step": 9310}, {"loss": 0.9198, "grad_norm": 1.2783401012420654, "learning_rate": 0.0002, "epoch": 6.091503267973856, "step": 9320}, {"loss": 0.8294, "grad_norm": 0.9916909337043762, "learning_rate": 0.0002, "epoch": 6.098039215686274, "step": 9330}, {"loss": 0.876, "grad_norm": 0.9300099015235901, "learning_rate": 0.0002, "epoch": 6.104575163398692, "step": 9340}, {"loss": 0.9064, "grad_norm": 1.4985264539718628, "learning_rate": 0.0002, "epoch": 6.111111111111111, "step": 9350}, {"loss": 1.0106, "grad_norm": 1.276380181312561, "learning_rate": 0.0002, "epoch": 6.117647058823529, "step": 9360}, {"loss": 0.9068, "grad_norm": 1.181113600730896, "learning_rate": 0.0002, "epoch": 6.124183006535947, "step": 9370}, {"loss": 0.9165, "grad_norm": 1.698729395866394, "learning_rate": 0.0002, "epoch": 6.130718954248366, "step": 9380}, {"loss": 0.7997, "grad_norm": 0.9793189764022827, "learning_rate": 0.0002, "epoch": 6.137254901960785, "step": 9390}, {"loss": 0.9731, "grad_norm": 1.1942132711410522, "learning_rate": 0.0002, "epoch": 6.143790849673203, "step": 9400}, {"loss": 0.8762, "grad_norm": 1.2160184383392334, "learning_rate": 0.0002, "epoch": 6.150326797385621, "step": 9410}, {"loss": 0.801, "grad_norm": 1.0802825689315796, "learning_rate": 0.0002, "epoch": 6.1568627450980395, "step": 9420}, {"loss": 0.9055, "grad_norm": 3.024529218673706, "learning_rate": 0.0002, "epoch": 6.163398692810458, "step": 9430}, {"loss": 0.8739, "grad_norm": 0.975062370300293, "learning_rate": 0.0002, "epoch": 6.169934640522876, "step": 9440}, {"loss": 0.8485, "grad_norm": 0.9243306517601013, "learning_rate": 0.0002, "epoch": 6.176470588235294, "step": 9450}, {"loss": 0.947, "grad_norm": 0.8892099857330322, "learning_rate": 0.0002, "epoch": 6.183006535947713, "step": 9460}, {"loss": 0.9165, "grad_norm": 1.4151731729507446, "learning_rate": 0.0002, "epoch": 6.189542483660131, "step": 9470}, {"loss": 1.022, "grad_norm": 1.064701795578003, "learning_rate": 0.0002, "epoch": 6.196078431372549, "step": 9480}, {"loss": 0.906, "grad_norm": 1.1104519367218018, "learning_rate": 0.0002, "epoch": 6.2026143790849675, "step": 9490}, {"loss": 0.9572, "grad_norm": 1.4788947105407715, "learning_rate": 0.0002, "epoch": 6.209150326797386, "step": 9500}, {"loss": 0.8014, "grad_norm": 0.7976077795028687, "learning_rate": 0.0002, "epoch": 6.215686274509804, "step": 9510}, {"loss": 0.886, "grad_norm": 1.256864070892334, "learning_rate": 0.0002, "epoch": 6.222222222222222, "step": 9520}, {"loss": 0.9104, "grad_norm": 1.3874554634094238, "learning_rate": 0.0002, "epoch": 6.228758169934641, "step": 9530}, {"loss": 0.8583, "grad_norm": 1.9012963771820068, "learning_rate": 0.0002, "epoch": 6.235294117647059, "step": 9540}, {"loss": 0.9585, "grad_norm": 1.275212287902832, "learning_rate": 0.0002, "epoch": 6.241830065359477, "step": 9550}, {"loss": 0.8416, "grad_norm": 1.1007417440414429, "learning_rate": 0.0002, "epoch": 6.248366013071895, "step": 9560}, {"loss": 0.9191, "grad_norm": 1.0602147579193115, "learning_rate": 0.0002, "epoch": 6.254901960784314, "step": 9570}, {"loss": 0.909, "grad_norm": 1.2276418209075928, "learning_rate": 0.0002, "epoch": 6.261437908496732, "step": 9580}, {"loss": 0.9363, "grad_norm": 1.0111924409866333, "learning_rate": 0.0002, "epoch": 6.26797385620915, "step": 9590}, {"loss": 0.9941, "grad_norm": 0.9031485915184021, "learning_rate": 0.0002, "epoch": 6.2745098039215685, "step": 9600}, {"loss": 0.9138, "grad_norm": 0.9893783926963806, "learning_rate": 0.0002, "epoch": 6.281045751633987, "step": 9610}, {"loss": 0.9114, "grad_norm": 1.1979725360870361, "learning_rate": 0.0002, "epoch": 6.287581699346405, "step": 9620}, {"loss": 0.8858, "grad_norm": 1.380516767501831, "learning_rate": 0.0002, "epoch": 6.294117647058823, "step": 9630}, {"loss": 0.8898, "grad_norm": 1.1370083093643188, "learning_rate": 0.0002, "epoch": 6.300653594771242, "step": 9640}, {"loss": 0.9073, "grad_norm": 1.4091558456420898, "learning_rate": 0.0002, "epoch": 6.30718954248366, "step": 9650}, {"loss": 0.9096, "grad_norm": 1.0670944452285767, "learning_rate": 0.0002, "epoch": 6.313725490196078, "step": 9660}, {"loss": 0.9376, "grad_norm": 0.9150263667106628, "learning_rate": 0.0002, "epoch": 6.3202614379084965, "step": 9670}, {"loss": 0.9169, "grad_norm": 1.1342853307724, "learning_rate": 0.0002, "epoch": 6.326797385620915, "step": 9680}, {"loss": 1.002, "grad_norm": 1.2733415365219116, "learning_rate": 0.0002, "epoch": 6.333333333333333, "step": 9690}, {"loss": 0.9579, "grad_norm": 1.3647292852401733, "learning_rate": 0.0002, "epoch": 6.339869281045751, "step": 9700}, {"loss": 0.87, "grad_norm": 1.0435094833374023, "learning_rate": 0.0002, "epoch": 6.34640522875817, "step": 9710}, {"loss": 0.8812, "grad_norm": 1.3641071319580078, "learning_rate": 0.0002, "epoch": 6.352941176470588, "step": 9720}, {"loss": 0.8888, "grad_norm": 1.2806159257888794, "learning_rate": 0.0002, "epoch": 6.359477124183006, "step": 9730}, {"loss": 0.9481, "grad_norm": 1.0193076133728027, "learning_rate": 0.0002, "epoch": 6.366013071895424, "step": 9740}, {"loss": 0.931, "grad_norm": 1.2349408864974976, "learning_rate": 0.0002, "epoch": 6.372549019607844, "step": 9750}, {"loss": 0.8837, "grad_norm": 1.2062549591064453, "learning_rate": 0.0002, "epoch": 6.379084967320262, "step": 9760}, {"loss": 0.8947, "grad_norm": 1.4402194023132324, "learning_rate": 0.0002, "epoch": 6.38562091503268, "step": 9770}, {"loss": 0.8724, "grad_norm": 1.1730891466140747, "learning_rate": 0.0002, "epoch": 6.392156862745098, "step": 9780}, {"loss": 0.9005, "grad_norm": 1.1481093168258667, "learning_rate": 0.0002, "epoch": 6.398692810457517, "step": 9790}, {"loss": 0.9431, "grad_norm": 1.0012723207473755, "learning_rate": 0.0002, "epoch": 6.405228758169935, "step": 9800}, {"loss": 0.8856, "grad_norm": 0.8839848041534424, "learning_rate": 0.0002, "epoch": 6.411764705882353, "step": 9810}, {"loss": 0.8147, "grad_norm": 1.096693992614746, "learning_rate": 0.0002, "epoch": 6.4183006535947715, "step": 9820}, {"loss": 0.846, "grad_norm": 1.4713369607925415, "learning_rate": 0.0002, "epoch": 6.42483660130719, "step": 9830}, {"loss": 0.9563, "grad_norm": 1.2529761791229248, "learning_rate": 0.0002, "epoch": 6.431372549019608, "step": 9840}, {"loss": 0.8551, "grad_norm": 1.5575600862503052, "learning_rate": 0.0002, "epoch": 6.437908496732026, "step": 9850}, {"loss": 0.836, "grad_norm": 1.2188916206359863, "learning_rate": 0.0002, "epoch": 6.444444444444445, "step": 9860}, {"loss": 0.9132, "grad_norm": 1.1558794975280762, "learning_rate": 0.0002, "epoch": 6.450980392156863, "step": 9870}, {"loss": 0.8632, "grad_norm": 1.1506937742233276, "learning_rate": 0.0002, "epoch": 6.457516339869281, "step": 9880}, {"loss": 1.0575, "grad_norm": 1.1168335676193237, "learning_rate": 0.0002, "epoch": 6.4640522875816995, "step": 9890}, {"loss": 0.99, "grad_norm": 1.192449688911438, "learning_rate": 0.0002, "epoch": 6.470588235294118, "step": 9900}, {"loss": 0.9478, "grad_norm": 1.0451104640960693, "learning_rate": 0.0002, "epoch": 6.477124183006536, "step": 9910}, {"loss": 0.9034, "grad_norm": 1.1111775636672974, "learning_rate": 0.0002, "epoch": 6.483660130718954, "step": 9920}, {"loss": 0.8971, "grad_norm": 1.2094531059265137, "learning_rate": 0.0002, "epoch": 6.490196078431373, "step": 9930}, {"loss": 0.9047, "grad_norm": 1.0547380447387695, "learning_rate": 0.0002, "epoch": 6.496732026143791, "step": 9940}, {"loss": 1.0727, "grad_norm": 1.5547202825546265, "learning_rate": 0.0002, "epoch": 6.503267973856209, "step": 9950}, {"loss": 0.9109, "grad_norm": 1.1917903423309326, "learning_rate": 0.0002, "epoch": 6.509803921568627, "step": 9960}, {"loss": 0.8708, "grad_norm": 1.0918153524398804, "learning_rate": 0.0002, "epoch": 6.516339869281046, "step": 9970}, {"loss": 0.8752, "grad_norm": 1.146968960762024, "learning_rate": 0.0002, "epoch": 6.522875816993464, "step": 9980}, {"loss": 0.9593, "grad_norm": 0.9899234771728516, "learning_rate": 0.0002, "epoch": 6.529411764705882, "step": 9990}, {"loss": 0.91, "grad_norm": 2.160924196243286, "learning_rate": 0.0002, "epoch": 6.5359477124183005, "step": 10000}, {"loss": 0.9683, "grad_norm": 1.6366891860961914, "learning_rate": 0.0002, "epoch": 6.542483660130719, "step": 10010}, {"loss": 0.8582, "grad_norm": 0.9876762628555298, "learning_rate": 0.0002, "epoch": 6.549019607843137, "step": 10020}, {"loss": 0.8385, "grad_norm": 1.5622549057006836, "learning_rate": 0.0002, "epoch": 6.555555555555555, "step": 10030}, {"loss": 0.8791, "grad_norm": 1.0108020305633545, "learning_rate": 0.0002, "epoch": 6.562091503267974, "step": 10040}, {"loss": 0.9574, "grad_norm": 1.0725725889205933, "learning_rate": 0.0002, "epoch": 6.568627450980392, "step": 10050}, {"loss": 0.8297, "grad_norm": 1.1551216840744019, "learning_rate": 0.0002, "epoch": 6.57516339869281, "step": 10060}, {"loss": 0.8199, "grad_norm": 1.5174646377563477, "learning_rate": 0.0002, "epoch": 6.5816993464052285, "step": 10070}, {"loss": 0.8203, "grad_norm": 1.041877031326294, "learning_rate": 0.0002, "epoch": 6.588235294117647, "step": 10080}, {"loss": 0.9684, "grad_norm": 0.9939621686935425, "learning_rate": 0.0002, "epoch": 6.594771241830065, "step": 10090}, {"loss": 0.9324, "grad_norm": 1.2706589698791504, "learning_rate": 0.0002, "epoch": 6.601307189542483, "step": 10100}, {"loss": 0.9614, "grad_norm": 1.1071467399597168, "learning_rate": 0.0002, "epoch": 6.607843137254902, "step": 10110}, {"loss": 0.9747, "grad_norm": 0.9449541568756104, "learning_rate": 0.0002, "epoch": 6.61437908496732, "step": 10120}, {"loss": 0.9557, "grad_norm": 1.0961830615997314, "learning_rate": 0.0002, "epoch": 6.620915032679738, "step": 10130}, {"loss": 0.9865, "grad_norm": 1.7726300954818726, "learning_rate": 0.0002, "epoch": 6.627450980392156, "step": 10140}, {"loss": 0.9657, "grad_norm": 1.2345516681671143, "learning_rate": 0.0002, "epoch": 6.633986928104575, "step": 10150}, {"loss": 0.9573, "grad_norm": 1.2062907218933105, "learning_rate": 0.0002, "epoch": 6.640522875816993, "step": 10160}, {"loss": 0.918, "grad_norm": 1.029327154159546, "learning_rate": 0.0002, "epoch": 6.647058823529412, "step": 10170}, {"loss": 0.9211, "grad_norm": 1.442307710647583, "learning_rate": 0.0002, "epoch": 6.65359477124183, "step": 10180}, {"loss": 0.8924, "grad_norm": 1.2579066753387451, "learning_rate": 0.0002, "epoch": 6.660130718954249, "step": 10190}, {"loss": 0.9836, "grad_norm": 1.4563188552856445, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 10200}, {"loss": 0.8876, "grad_norm": 0.9699450135231018, "learning_rate": 0.0002, "epoch": 6.673202614379085, "step": 10210}, {"loss": 0.9589, "grad_norm": 1.812523603439331, "learning_rate": 0.0002, "epoch": 6.6797385620915035, "step": 10220}, {"loss": 1.0241, "grad_norm": 1.124000906944275, "learning_rate": 0.0002, "epoch": 6.686274509803922, "step": 10230}, {"loss": 0.8924, "grad_norm": 1.0957475900650024, "learning_rate": 0.0002, "epoch": 6.69281045751634, "step": 10240}, {"loss": 0.8891, "grad_norm": 0.989689826965332, "learning_rate": 0.0002, "epoch": 6.699346405228758, "step": 10250}, {"loss": 0.9049, "grad_norm": 1.4353317022323608, "learning_rate": 0.0002, "epoch": 6.705882352941177, "step": 10260}, {"loss": 0.9311, "grad_norm": 1.0245451927185059, "learning_rate": 0.0002, "epoch": 6.712418300653595, "step": 10270}, {"loss": 0.8814, "grad_norm": 1.097334861755371, "learning_rate": 0.0002, "epoch": 6.718954248366013, "step": 10280}, {"loss": 0.9927, "grad_norm": 0.982356071472168, "learning_rate": 0.0002, "epoch": 6.7254901960784315, "step": 10290}, {"loss": 0.9909, "grad_norm": 1.8842819929122925, "learning_rate": 0.0002, "epoch": 6.73202614379085, "step": 10300}, {"loss": 0.9286, "grad_norm": 0.8648947477340698, "learning_rate": 0.0002, "epoch": 6.738562091503268, "step": 10310}, {"loss": 0.987, "grad_norm": 1.1510577201843262, "learning_rate": 0.0002, "epoch": 6.745098039215686, "step": 10320}, {"loss": 0.9217, "grad_norm": 1.874495506286621, "learning_rate": 0.0002, "epoch": 6.751633986928105, "step": 10330}, {"loss": 0.8914, "grad_norm": 1.1126408576965332, "learning_rate": 0.0002, "epoch": 6.758169934640523, "step": 10340}, {"loss": 0.8508, "grad_norm": 1.6654644012451172, "learning_rate": 0.0002, "epoch": 6.764705882352941, "step": 10350}, {"loss": 0.9653, "grad_norm": 1.0699580907821655, "learning_rate": 0.0002, "epoch": 6.771241830065359, "step": 10360}, {"loss": 0.882, "grad_norm": 0.9460757374763489, "learning_rate": 0.0002, "epoch": 6.777777777777778, "step": 10370}, {"loss": 0.9589, "grad_norm": 1.2553058862686157, "learning_rate": 0.0002, "epoch": 6.784313725490196, "step": 10380}, {"loss": 0.8782, "grad_norm": 1.0939891338348389, "learning_rate": 0.0002, "epoch": 6.790849673202614, "step": 10390}, {"loss": 0.9189, "grad_norm": 1.0647451877593994, "learning_rate": 0.0002, "epoch": 6.7973856209150325, "step": 10400}, {"loss": 0.9478, "grad_norm": 1.0954521894454956, "learning_rate": 0.0002, "epoch": 6.803921568627451, "step": 10410}, {"loss": 1.0385, "grad_norm": 1.4371392726898193, "learning_rate": 0.0002, "epoch": 6.810457516339869, "step": 10420}, {"loss": 1.0024, "grad_norm": 1.0063464641571045, "learning_rate": 0.0002, "epoch": 6.816993464052287, "step": 10430}, {"loss": 0.8737, "grad_norm": 1.5189263820648193, "learning_rate": 0.0002, "epoch": 6.823529411764706, "step": 10440}, {"loss": 0.9246, "grad_norm": 0.9715501070022583, "learning_rate": 0.0002, "epoch": 6.830065359477124, "step": 10450}, {"loss": 0.9659, "grad_norm": 1.114586353302002, "learning_rate": 0.0002, "epoch": 6.836601307189542, "step": 10460}, {"loss": 1.0081, "grad_norm": 1.2991431951522827, "learning_rate": 0.0002, "epoch": 6.8431372549019605, "step": 10470}, {"loss": 0.9323, "grad_norm": 1.203114628791809, "learning_rate": 0.0002, "epoch": 6.849673202614379, "step": 10480}, {"loss": 1.0032, "grad_norm": 1.476167917251587, "learning_rate": 0.0002, "epoch": 6.856209150326797, "step": 10490}, {"loss": 1.0275, "grad_norm": 1.0933326482772827, "learning_rate": 0.0002, "epoch": 6.862745098039216, "step": 10500}, {"loss": 1.0068, "grad_norm": 1.2831504344940186, "learning_rate": 0.0002, "epoch": 6.8692810457516345, "step": 10510}, {"loss": 0.9973, "grad_norm": 1.1967637538909912, "learning_rate": 0.0002, "epoch": 6.875816993464053, "step": 10520}, {"loss": 0.9549, "grad_norm": 1.1276888847351074, "learning_rate": 0.0002, "epoch": 6.882352941176471, "step": 10530}, {"loss": 0.9568, "grad_norm": 1.2680490016937256, "learning_rate": 0.0002, "epoch": 6.888888888888889, "step": 10540}, {"loss": 0.9177, "grad_norm": 1.5469038486480713, "learning_rate": 0.0002, "epoch": 6.895424836601308, "step": 10550}, {"loss": 0.8545, "grad_norm": 1.1731038093566895, "learning_rate": 0.0002, "epoch": 6.901960784313726, "step": 10560}, {"loss": 0.9795, "grad_norm": 0.968008816242218, "learning_rate": 0.0002, "epoch": 6.908496732026144, "step": 10570}, {"loss": 0.9439, "grad_norm": 0.9082416892051697, "learning_rate": 0.0002, "epoch": 6.915032679738562, "step": 10580}, {"loss": 0.9898, "grad_norm": 1.5816899538040161, "learning_rate": 0.0002, "epoch": 6.921568627450981, "step": 10590}, {"loss": 0.9692, "grad_norm": 0.9462234377861023, "learning_rate": 0.0002, "epoch": 6.928104575163399, "step": 10600}, {"loss": 1.0193, "grad_norm": 1.4950200319290161, "learning_rate": 0.0002, "epoch": 6.934640522875817, "step": 10610}, {"loss": 0.8888, "grad_norm": 1.2929182052612305, "learning_rate": 0.0002, "epoch": 6.9411764705882355, "step": 10620}, {"loss": 1.0141, "grad_norm": 1.2995754480361938, "learning_rate": 0.0002, "epoch": 6.947712418300654, "step": 10630}, {"loss": 0.9863, "grad_norm": 0.9407122135162354, "learning_rate": 0.0002, "epoch": 6.954248366013072, "step": 10640}, {"loss": 0.9041, "grad_norm": 1.1735378503799438, "learning_rate": 0.0002, "epoch": 6.96078431372549, "step": 10650}, {"loss": 0.936, "grad_norm": 0.9937344193458557, "learning_rate": 0.0002, "epoch": 6.967320261437909, "step": 10660}, {"loss": 0.9577, "grad_norm": 1.2498728036880493, "learning_rate": 0.0002, "epoch": 6.973856209150327, "step": 10670}, {"loss": 1.0504, "grad_norm": 1.0513341426849365, "learning_rate": 0.0002, "epoch": 6.980392156862745, "step": 10680}, {"loss": 0.9259, "grad_norm": 1.4611467123031616, "learning_rate": 0.0002, "epoch": 6.9869281045751634, "step": 10690}, {"loss": 0.9779, "grad_norm": 1.2924799919128418, "learning_rate": 0.0002, "epoch": 6.993464052287582, "step": 10700}, {"loss": 0.8953, "grad_norm": 1.2024929523468018, "learning_rate": 0.0002, "epoch": 7.0, "step": 10710}]}
+{"epoch": 8.0, "step": 12240, "epoch_duration": 1811.3724415302277, "total_accumulated_duration": 13780.879077911377, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 3020.60888671875}, "peak_memory_usage": {"GPU_0": 15051.1748046875}, "avg_memory_reserved": {"GPU_0": 15256.0}, "peak_memory_reserved": {"GPU_0": 16176.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-9809-sd-1/checkpoint-6120", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 4.7451, "grad_norm": 1.5105072259902954, "learning_rate": 0.0002, "epoch": 0.006535947712418301, "step": 10}, {"loss": 3.3158, "grad_norm": 2.1156165599823, "learning_rate": 0.0002, "epoch": 0.013071895424836602, "step": 20}, {"loss": 2.643, "grad_norm": 1.0578808784484863, "learning_rate": 0.0002, "epoch": 0.0196078431372549, "step": 30}, {"loss": 2.3948, "grad_norm": 2.725064516067505, "learning_rate": 0.0002, "epoch": 0.026143790849673203, "step": 40}, {"loss": 2.3134, "grad_norm": 2.9575750827789307, "learning_rate": 0.0002, "epoch": 0.032679738562091505, "step": 50}, {"loss": 2.2778, "grad_norm": 1.2158117294311523, "learning_rate": 0.0002, "epoch": 0.0392156862745098, "step": 60}, {"loss": 1.9742, "grad_norm": 1.0850954055786133, "learning_rate": 0.0002, "epoch": 0.0457516339869281, "step": 70}, {"loss": 1.8872, "grad_norm": 1.299196720123291, "learning_rate": 0.0002, "epoch": 0.05228758169934641, "step": 80}, {"loss": 1.947, "grad_norm": 0.8310191035270691, "learning_rate": 0.0002, "epoch": 0.058823529411764705, "step": 90}, {"loss": 1.9098, "grad_norm": 0.9854435920715332, "learning_rate": 0.0002, "epoch": 0.06535947712418301, "step": 100}, {"loss": 1.7508, "grad_norm": 0.7951157689094543, "learning_rate": 0.0002, "epoch": 0.0718954248366013, "step": 110}, {"loss": 1.9035, "grad_norm": 0.7593062520027161, "learning_rate": 0.0002, "epoch": 0.0784313725490196, "step": 120}, {"loss": 1.8517, "grad_norm": 0.6783032417297363, "learning_rate": 0.0002, "epoch": 0.08496732026143791, "step": 130}, {"loss": 1.6805, "grad_norm": 0.8350756764411926, "learning_rate": 0.0002, "epoch": 0.0915032679738562, "step": 140}, {"loss": 1.6123, "grad_norm": 1.0203173160552979, "learning_rate": 0.0002, "epoch": 0.09803921568627451, "step": 150}, {"loss": 1.7248, "grad_norm": 0.8820539712905884, "learning_rate": 0.0002, "epoch": 0.10457516339869281, "step": 160}, {"loss": 1.6762, "grad_norm": 0.7286128997802734, "learning_rate": 0.0002, "epoch": 0.1111111111111111, "step": 170}, {"loss": 1.8841, "grad_norm": 0.7874041795730591, "learning_rate": 0.0002, "epoch": 0.11764705882352941, "step": 180}, {"loss": 1.5656, "grad_norm": 0.6630475521087646, "learning_rate": 0.0002, "epoch": 0.12418300653594772, "step": 190}, {"loss": 1.6149, "grad_norm": 0.686413586139679, "learning_rate": 0.0002, "epoch": 0.13071895424836602, "step": 200}, {"loss": 1.6227, "grad_norm": 0.7793629765510559, "learning_rate": 0.0002, "epoch": 0.13725490196078433, "step": 210}, {"loss": 1.7223, "grad_norm": 0.6893141865730286, "learning_rate": 0.0002, "epoch": 0.1437908496732026, "step": 220}, {"loss": 1.6808, "grad_norm": 0.5804724097251892, "learning_rate": 0.0002, "epoch": 0.1503267973856209, "step": 230}, {"loss": 1.5578, "grad_norm": 0.6053574085235596, "learning_rate": 0.0002, "epoch": 0.1568627450980392, "step": 240}, {"loss": 1.7394, "grad_norm": 0.7566025853157043, "learning_rate": 0.0002, "epoch": 0.16339869281045752, "step": 250}, {"loss": 1.6216, "grad_norm": 0.6112990975379944, "learning_rate": 0.0002, "epoch": 0.16993464052287582, "step": 260}, {"loss": 1.5564, "grad_norm": 0.6839066743850708, "learning_rate": 0.0002, "epoch": 0.17647058823529413, "step": 270}, {"loss": 1.7129, "grad_norm": 0.6368117928504944, "learning_rate": 0.0002, "epoch": 0.1830065359477124, "step": 280}, {"loss": 1.5646, "grad_norm": 0.6144475936889648, "learning_rate": 0.0002, "epoch": 0.1895424836601307, "step": 290}, {"loss": 1.8383, "grad_norm": 0.6743767261505127, "learning_rate": 0.0002, "epoch": 0.19607843137254902, "step": 300}, {"loss": 1.421, "grad_norm": 0.6807955503463745, "learning_rate": 0.0002, "epoch": 0.20261437908496732, "step": 310}, {"loss": 1.5961, "grad_norm": 0.6717963814735413, "learning_rate": 0.0002, "epoch": 0.20915032679738563, "step": 320}, {"loss": 1.6842, "grad_norm": 0.5917780995368958, "learning_rate": 0.0002, "epoch": 0.21568627450980393, "step": 330}, {"loss": 1.6264, "grad_norm": 0.6783658862113953, "learning_rate": 0.0002, "epoch": 0.2222222222222222, "step": 340}, {"loss": 1.4635, "grad_norm": 0.5820256471633911, "learning_rate": 0.0002, "epoch": 0.22875816993464052, "step": 350}, {"loss": 1.6514, "grad_norm": 0.5345938801765442, "learning_rate": 0.0002, "epoch": 0.23529411764705882, "step": 360}, {"loss": 1.6441, "grad_norm": 0.755929172039032, "learning_rate": 0.0002, "epoch": 0.24183006535947713, "step": 370}, {"loss": 1.5177, "grad_norm": 0.6183189749717712, "learning_rate": 0.0002, "epoch": 0.24836601307189543, "step": 380}, {"loss": 1.5935, "grad_norm": 0.7277782559394836, "learning_rate": 0.0002, "epoch": 0.2549019607843137, "step": 390}, {"loss": 1.6957, "grad_norm": 0.9998756051063538, "learning_rate": 0.0002, "epoch": 0.26143790849673204, "step": 400}, {"loss": 1.5738, "grad_norm": 0.7523853778839111, "learning_rate": 0.0002, "epoch": 0.2679738562091503, "step": 410}, {"loss": 1.5649, "grad_norm": 0.6548714637756348, "learning_rate": 0.0002, "epoch": 0.27450980392156865, "step": 420}, {"loss": 1.4564, "grad_norm": 0.6979796290397644, "learning_rate": 0.0002, "epoch": 0.28104575163398693, "step": 430}, {"loss": 1.5927, "grad_norm": 0.840915322303772, "learning_rate": 0.0002, "epoch": 0.2875816993464052, "step": 440}, {"loss": 1.5199, "grad_norm": 0.6142978072166443, "learning_rate": 0.0002, "epoch": 0.29411764705882354, "step": 450}, {"loss": 1.4903, "grad_norm": 0.9482691884040833, "learning_rate": 0.0002, "epoch": 0.3006535947712418, "step": 460}, {"loss": 1.6553, "grad_norm": 0.7001156806945801, "learning_rate": 0.0002, "epoch": 0.30718954248366015, "step": 470}, {"loss": 1.5957, "grad_norm": 0.6665455102920532, "learning_rate": 0.0002, "epoch": 0.3137254901960784, "step": 480}, {"loss": 1.587, "grad_norm": 0.6012697815895081, "learning_rate": 0.0002, "epoch": 0.3202614379084967, "step": 490}, {"loss": 1.4468, "grad_norm": 0.8770062327384949, "learning_rate": 0.0002, "epoch": 0.32679738562091504, "step": 500}, {"loss": 1.3558, "grad_norm": 0.7029962539672852, "learning_rate": 0.0002, "epoch": 0.3333333333333333, "step": 510}, {"loss": 1.4435, "grad_norm": 0.6682832837104797, "learning_rate": 0.0002, "epoch": 0.33986928104575165, "step": 520}, {"loss": 1.4242, "grad_norm": 0.5548969507217407, "learning_rate": 0.0002, "epoch": 0.3464052287581699, "step": 530}, {"loss": 1.5081, "grad_norm": 0.6640702486038208, "learning_rate": 0.0002, "epoch": 0.35294117647058826, "step": 540}, {"loss": 1.4998, "grad_norm": 0.656292200088501, "learning_rate": 0.0002, "epoch": 0.35947712418300654, "step": 550}, {"loss": 1.5415, "grad_norm": 0.618910551071167, "learning_rate": 0.0002, "epoch": 0.3660130718954248, "step": 560}, {"loss": 1.5178, "grad_norm": 0.644859790802002, "learning_rate": 0.0002, "epoch": 0.37254901960784315, "step": 570}, {"loss": 1.645, "grad_norm": 0.679042398929596, "learning_rate": 0.0002, "epoch": 0.3790849673202614, "step": 580}, {"loss": 1.5193, "grad_norm": 0.980681836605072, "learning_rate": 0.0002, "epoch": 0.38562091503267976, "step": 590}, {"loss": 1.4262, "grad_norm": 0.632219672203064, "learning_rate": 0.0002, "epoch": 0.39215686274509803, "step": 600}, {"loss": 1.5533, "grad_norm": 0.7003744840621948, "learning_rate": 0.0002, "epoch": 0.39869281045751637, "step": 610}, {"loss": 1.7747, "grad_norm": 0.7090577483177185, "learning_rate": 0.0002, "epoch": 0.40522875816993464, "step": 620}, {"loss": 1.7506, "grad_norm": 0.657819926738739, "learning_rate": 0.0002, "epoch": 0.4117647058823529, "step": 630}, {"loss": 1.621, "grad_norm": 0.7034208178520203, "learning_rate": 0.0002, "epoch": 0.41830065359477125, "step": 640}, {"loss": 1.5357, "grad_norm": 0.7274866104125977, "learning_rate": 0.0002, "epoch": 0.42483660130718953, "step": 650}, {"loss": 1.6304, "grad_norm": 0.5876233577728271, "learning_rate": 0.0002, "epoch": 0.43137254901960786, "step": 660}, {"loss": 1.7683, "grad_norm": 0.595494270324707, "learning_rate": 0.0002, "epoch": 0.43790849673202614, "step": 670}, {"loss": 1.5117, "grad_norm": 0.8253804445266724, "learning_rate": 0.0002, "epoch": 0.4444444444444444, "step": 680}, {"loss": 1.5199, "grad_norm": 0.652225911617279, "learning_rate": 0.0002, "epoch": 0.45098039215686275, "step": 690}, {"loss": 1.5419, "grad_norm": 0.6242014169692993, "learning_rate": 0.0002, "epoch": 0.45751633986928103, "step": 700}, {"loss": 1.53, "grad_norm": 0.7283986210823059, "learning_rate": 0.0002, "epoch": 0.46405228758169936, "step": 710}, {"loss": 1.43, "grad_norm": 0.7016081213951111, "learning_rate": 0.0002, "epoch": 0.47058823529411764, "step": 720}, {"loss": 1.4626, "grad_norm": 0.5211893916130066, "learning_rate": 0.0002, "epoch": 0.477124183006536, "step": 730}, {"loss": 1.6885, "grad_norm": 0.6221150159835815, "learning_rate": 0.0002, "epoch": 0.48366013071895425, "step": 740}, {"loss": 1.5677, "grad_norm": 0.76594477891922, "learning_rate": 0.0002, "epoch": 0.49019607843137253, "step": 750}, {"loss": 1.4982, "grad_norm": 0.5777859091758728, "learning_rate": 0.0002, "epoch": 0.49673202614379086, "step": 760}, {"loss": 1.5253, "grad_norm": 0.5793519616127014, "learning_rate": 0.0002, "epoch": 0.5032679738562091, "step": 770}, {"loss": 1.3562, "grad_norm": 0.5425786375999451, "learning_rate": 0.0002, "epoch": 0.5098039215686274, "step": 780}, {"loss": 1.3398, "grad_norm": 0.6004197001457214, "learning_rate": 0.0002, "epoch": 0.5163398692810458, "step": 790}, {"loss": 1.5346, "grad_norm": 0.7167016863822937, "learning_rate": 0.0002, "epoch": 0.5228758169934641, "step": 800}, {"loss": 1.48, "grad_norm": 0.710218071937561, "learning_rate": 0.0002, "epoch": 0.5294117647058824, "step": 810}, {"loss": 1.3943, "grad_norm": 0.699528694152832, "learning_rate": 0.0002, "epoch": 0.5359477124183006, "step": 820}, {"loss": 1.6014, "grad_norm": 0.579629123210907, "learning_rate": 0.0002, "epoch": 0.5424836601307189, "step": 830}, {"loss": 1.3894, "grad_norm": 0.595407247543335, "learning_rate": 0.0002, "epoch": 0.5490196078431373, "step": 840}, {"loss": 1.6394, "grad_norm": 0.544563889503479, "learning_rate": 0.0002, "epoch": 0.5555555555555556, "step": 850}, {"loss": 1.4692, "grad_norm": 0.553166389465332, "learning_rate": 0.0002, "epoch": 0.5620915032679739, "step": 860}, {"loss": 1.5155, "grad_norm": 0.5645018815994263, "learning_rate": 0.0002, "epoch": 0.5686274509803921, "step": 870}, {"loss": 1.7019, "grad_norm": 0.6576932668685913, "learning_rate": 0.0002, "epoch": 0.5751633986928104, "step": 880}, {"loss": 1.5891, "grad_norm": 0.6684197187423706, "learning_rate": 0.0002, "epoch": 0.5816993464052288, "step": 890}, {"loss": 1.5348, "grad_norm": 0.6706975698471069, "learning_rate": 0.0002, "epoch": 0.5882352941176471, "step": 900}, {"loss": 1.4038, "grad_norm": 0.6762327551841736, "learning_rate": 0.0002, "epoch": 0.5947712418300654, "step": 910}, {"loss": 1.61, "grad_norm": 0.764032244682312, "learning_rate": 0.0002, "epoch": 0.6013071895424836, "step": 920}, {"loss": 1.436, "grad_norm": 0.6996400952339172, "learning_rate": 0.0002, "epoch": 0.6078431372549019, "step": 930}, {"loss": 1.6038, "grad_norm": 0.686735987663269, "learning_rate": 0.0002, "epoch": 0.6143790849673203, "step": 940}, {"loss": 1.5194, "grad_norm": 0.6086131930351257, "learning_rate": 0.0002, "epoch": 0.6209150326797386, "step": 950}, {"loss": 1.4457, "grad_norm": 0.5627856850624084, "learning_rate": 0.0002, "epoch": 0.6274509803921569, "step": 960}, {"loss": 1.506, "grad_norm": 0.5781503319740295, "learning_rate": 0.0002, "epoch": 0.6339869281045751, "step": 970}, {"loss": 1.5668, "grad_norm": 0.6347246766090393, "learning_rate": 0.0002, "epoch": 0.6405228758169934, "step": 980}, {"loss": 1.3819, "grad_norm": 0.6581300497055054, "learning_rate": 0.0002, "epoch": 0.6470588235294118, "step": 990}, {"loss": 1.6425, "grad_norm": 0.8343676924705505, "learning_rate": 0.0002, "epoch": 0.6535947712418301, "step": 1000}, {"loss": 1.5188, "grad_norm": 0.5708910226821899, "learning_rate": 0.0002, "epoch": 0.6601307189542484, "step": 1010}, {"loss": 1.3882, "grad_norm": 0.6832585334777832, "learning_rate": 0.0002, "epoch": 0.6666666666666666, "step": 1020}, {"loss": 1.645, "grad_norm": 0.5767837166786194, "learning_rate": 0.0002, "epoch": 0.673202614379085, "step": 1030}, {"loss": 1.4206, "grad_norm": 0.5637745261192322, "learning_rate": 0.0002, "epoch": 0.6797385620915033, "step": 1040}, {"loss": 1.4325, "grad_norm": 0.8193050026893616, "learning_rate": 0.0002, "epoch": 0.6862745098039216, "step": 1050}, {"loss": 1.4196, "grad_norm": 0.6157439351081848, "learning_rate": 0.0002, "epoch": 0.6928104575163399, "step": 1060}, {"loss": 1.5547, "grad_norm": 0.7476664781570435, "learning_rate": 0.0002, "epoch": 0.6993464052287581, "step": 1070}, {"loss": 1.5337, "grad_norm": 0.8569361567497253, "learning_rate": 0.0002, "epoch": 0.7058823529411765, "step": 1080}, {"loss": 1.482, "grad_norm": 0.5671911835670471, "learning_rate": 0.0002, "epoch": 0.7124183006535948, "step": 1090}, {"loss": 1.5398, "grad_norm": 0.5151128768920898, "learning_rate": 0.0002, "epoch": 0.7189542483660131, "step": 1100}, {"loss": 1.4848, "grad_norm": 0.568037211894989, "learning_rate": 0.0002, "epoch": 0.7254901960784313, "step": 1110}, {"loss": 1.4708, "grad_norm": 0.6756396889686584, "learning_rate": 0.0002, "epoch": 0.7320261437908496, "step": 1120}, {"loss": 1.4017, "grad_norm": 0.638975977897644, "learning_rate": 0.0002, "epoch": 0.738562091503268, "step": 1130}, {"loss": 1.6028, "grad_norm": 0.7103341221809387, "learning_rate": 0.0002, "epoch": 0.7450980392156863, "step": 1140}, {"loss": 1.3766, "grad_norm": 0.7403952479362488, "learning_rate": 0.0002, "epoch": 0.7516339869281046, "step": 1150}, {"loss": 1.4757, "grad_norm": 0.6266511082649231, "learning_rate": 0.0002, "epoch": 0.7581699346405228, "step": 1160}, {"loss": 1.4468, "grad_norm": 0.5939070582389832, "learning_rate": 0.0002, "epoch": 0.7647058823529411, "step": 1170}, {"loss": 1.4145, "grad_norm": 0.5735430717468262, "learning_rate": 0.0002, "epoch": 0.7712418300653595, "step": 1180}, {"loss": 1.3891, "grad_norm": 0.5155234932899475, "learning_rate": 0.0002, "epoch": 0.7777777777777778, "step": 1190}, {"loss": 1.4942, "grad_norm": 0.5115423202514648, "learning_rate": 0.0002, "epoch": 0.7843137254901961, "step": 1200}, {"loss": 1.4508, "grad_norm": 0.693588137626648, "learning_rate": 0.0002, "epoch": 0.7908496732026143, "step": 1210}, {"loss": 1.308, "grad_norm": 0.5504693984985352, "learning_rate": 0.0002, "epoch": 0.7973856209150327, "step": 1220}, {"loss": 1.5412, "grad_norm": 0.5555992126464844, "learning_rate": 0.0002, "epoch": 0.803921568627451, "step": 1230}, {"loss": 1.5506, "grad_norm": 0.7211785316467285, "learning_rate": 0.0002, "epoch": 0.8104575163398693, "step": 1240}, {"loss": 1.6163, "grad_norm": 0.735003650188446, "learning_rate": 0.0002, "epoch": 0.8169934640522876, "step": 1250}, {"loss": 1.5836, "grad_norm": 0.5245152711868286, "learning_rate": 0.0002, "epoch": 0.8235294117647058, "step": 1260}, {"loss": 1.4505, "grad_norm": 0.5883445739746094, "learning_rate": 0.0002, "epoch": 0.8300653594771242, "step": 1270}, {"loss": 1.3642, "grad_norm": 0.6835859417915344, "learning_rate": 0.0002, "epoch": 0.8366013071895425, "step": 1280}, {"loss": 1.5526, "grad_norm": 0.6592142581939697, "learning_rate": 0.0002, "epoch": 0.8431372549019608, "step": 1290}, {"loss": 1.52, "grad_norm": 0.6087474226951599, "learning_rate": 0.0002, "epoch": 0.8496732026143791, "step": 1300}, {"loss": 1.3807, "grad_norm": 0.565387487411499, "learning_rate": 0.0002, "epoch": 0.8562091503267973, "step": 1310}, {"loss": 1.4809, "grad_norm": 0.7363151907920837, "learning_rate": 0.0002, "epoch": 0.8627450980392157, "step": 1320}, {"loss": 1.5683, "grad_norm": 0.5964524149894714, "learning_rate": 0.0002, "epoch": 0.869281045751634, "step": 1330}, {"loss": 1.3284, "grad_norm": 0.5169979929924011, "learning_rate": 0.0002, "epoch": 0.8758169934640523, "step": 1340}, {"loss": 1.6279, "grad_norm": 0.7063422799110413, "learning_rate": 0.0002, "epoch": 0.8823529411764706, "step": 1350}, {"loss": 1.3072, "grad_norm": 0.7261926531791687, "learning_rate": 0.0002, "epoch": 0.8888888888888888, "step": 1360}, {"loss": 1.3619, "grad_norm": 0.6759744882583618, "learning_rate": 0.0002, "epoch": 0.8954248366013072, "step": 1370}, {"loss": 1.4079, "grad_norm": 0.675051212310791, "learning_rate": 0.0002, "epoch": 0.9019607843137255, "step": 1380}, {"loss": 1.6606, "grad_norm": 0.5613595843315125, "learning_rate": 0.0002, "epoch": 0.9084967320261438, "step": 1390}, {"loss": 1.414, "grad_norm": 0.611732006072998, "learning_rate": 0.0002, "epoch": 0.9150326797385621, "step": 1400}, {"loss": 1.5766, "grad_norm": 0.6365187168121338, "learning_rate": 0.0002, "epoch": 0.9215686274509803, "step": 1410}, {"loss": 1.7832, "grad_norm": 0.7810426354408264, "learning_rate": 0.0002, "epoch": 0.9281045751633987, "step": 1420}, {"loss": 1.5377, "grad_norm": 0.593891441822052, "learning_rate": 0.0002, "epoch": 0.934640522875817, "step": 1430}, {"loss": 1.4468, "grad_norm": 0.761585533618927, "learning_rate": 0.0002, "epoch": 0.9411764705882353, "step": 1440}, {"loss": 1.589, "grad_norm": 0.6114464998245239, "learning_rate": 0.0002, "epoch": 0.9477124183006536, "step": 1450}, {"loss": 1.4973, "grad_norm": 0.601044774055481, "learning_rate": 0.0002, "epoch": 0.954248366013072, "step": 1460}, {"loss": 1.4162, "grad_norm": 0.5484876036643982, "learning_rate": 0.0002, "epoch": 0.9607843137254902, "step": 1470}, {"loss": 1.4825, "grad_norm": 0.5383428335189819, "learning_rate": 0.0002, "epoch": 0.9673202614379085, "step": 1480}, {"loss": 1.5543, "grad_norm": 0.648106575012207, "learning_rate": 0.0002, "epoch": 0.9738562091503268, "step": 1490}, {"loss": 1.3638, "grad_norm": 0.6847249865531921, "learning_rate": 0.0002, "epoch": 0.9803921568627451, "step": 1500}, {"loss": 1.4247, "grad_norm": 0.6361058354377747, "learning_rate": 0.0002, "epoch": 0.9869281045751634, "step": 1510}, {"loss": 1.5131, "grad_norm": 0.646392285823822, "learning_rate": 0.0002, "epoch": 0.9934640522875817, "step": 1520}, {"loss": 1.3738, "grad_norm": 0.5391159057617188, "learning_rate": 0.0002, "epoch": 1.0, "step": 1530}, {"eval_loss": 1.4715123176574707, "eval_runtime": 30.5701, "eval_samples_per_second": 14.262, "eval_steps_per_second": 1.799, "epoch": 1.0, "step": 1530}, {"loss": 1.4827, "grad_norm": 0.5468988418579102, "learning_rate": 0.0002, "epoch": 1.0065359477124183, "step": 1540}, {"loss": 1.4342, "grad_norm": 0.629940927028656, "learning_rate": 0.0002, "epoch": 1.0130718954248366, "step": 1550}, {"loss": 1.4259, "grad_norm": 0.6411303281784058, "learning_rate": 0.0002, "epoch": 1.0196078431372548, "step": 1560}, {"loss": 1.3924, "grad_norm": 0.5619024038314819, "learning_rate": 0.0002, "epoch": 1.026143790849673, "step": 1570}, {"loss": 1.6086, "grad_norm": 0.6093462705612183, "learning_rate": 0.0002, "epoch": 1.0326797385620916, "step": 1580}, {"loss": 1.4547, "grad_norm": 0.5543286204338074, "learning_rate": 0.0002, "epoch": 1.0392156862745099, "step": 1590}, {"loss": 1.3738, "grad_norm": 0.6079006195068359, "learning_rate": 0.0002, "epoch": 1.0457516339869282, "step": 1600}, {"loss": 1.4574, "grad_norm": 0.6240813136100769, "learning_rate": 0.0002, "epoch": 1.0522875816993464, "step": 1610}, {"loss": 1.3504, "grad_norm": 0.6141977310180664, "learning_rate": 0.0002, "epoch": 1.0588235294117647, "step": 1620}, {"loss": 1.3668, "grad_norm": 0.5920178294181824, "learning_rate": 0.0002, "epoch": 1.065359477124183, "step": 1630}, {"loss": 1.3204, "grad_norm": 0.47620782256126404, "learning_rate": 0.0002, "epoch": 1.0718954248366013, "step": 1640}, {"loss": 1.3249, "grad_norm": 0.6826292872428894, "learning_rate": 0.0002, "epoch": 1.0784313725490196, "step": 1650}, {"loss": 1.2285, "grad_norm": 0.6182006597518921, "learning_rate": 0.0002, "epoch": 1.0849673202614378, "step": 1660}, {"loss": 1.2907, "grad_norm": 0.57639479637146, "learning_rate": 0.0002, "epoch": 1.091503267973856, "step": 1670}, {"loss": 1.4575, "grad_norm": 0.6696860194206238, "learning_rate": 0.0002, "epoch": 1.0980392156862746, "step": 1680}, {"loss": 1.4104, "grad_norm": 0.699221670627594, "learning_rate": 0.0002, "epoch": 1.1045751633986929, "step": 1690}, {"loss": 1.3667, "grad_norm": 0.7138059139251709, "learning_rate": 0.0002, "epoch": 1.1111111111111112, "step": 1700}, {"loss": 1.3468, "grad_norm": 0.6930422186851501, "learning_rate": 0.0002, "epoch": 1.1176470588235294, "step": 1710}, {"loss": 1.5033, "grad_norm": 0.7484048008918762, "learning_rate": 0.0002, "epoch": 1.1241830065359477, "step": 1720}, {"loss": 1.4582, "grad_norm": 0.5820090174674988, "learning_rate": 0.0002, "epoch": 1.130718954248366, "step": 1730}, {"loss": 1.3704, "grad_norm": 0.7143406867980957, "learning_rate": 0.0002, "epoch": 1.1372549019607843, "step": 1740}, {"loss": 1.277, "grad_norm": 0.5597584247589111, "learning_rate": 0.0002, "epoch": 1.1437908496732025, "step": 1750}, {"loss": 1.5403, "grad_norm": 0.5171173214912415, "learning_rate": 0.0002, "epoch": 1.1503267973856208, "step": 1760}, {"loss": 1.419, "grad_norm": 0.5951920747756958, "learning_rate": 0.0002, "epoch": 1.156862745098039, "step": 1770}, {"loss": 1.2929, "grad_norm": 0.7506247758865356, "learning_rate": 0.0002, "epoch": 1.1633986928104576, "step": 1780}, {"loss": 1.5475, "grad_norm": 0.5936487913131714, "learning_rate": 0.0002, "epoch": 1.1699346405228759, "step": 1790}, {"loss": 1.3567, "grad_norm": 0.688450038433075, "learning_rate": 0.0002, "epoch": 1.1764705882352942, "step": 1800}, {"loss": 1.314, "grad_norm": 0.671623170375824, "learning_rate": 0.0002, "epoch": 1.1830065359477124, "step": 1810}, {"loss": 1.3803, "grad_norm": 0.6911860704421997, "learning_rate": 0.0002, "epoch": 1.1895424836601307, "step": 1820}, {"loss": 1.363, "grad_norm": 0.60726398229599, "learning_rate": 0.0002, "epoch": 1.196078431372549, "step": 1830}, {"loss": 1.5236, "grad_norm": 0.7542088627815247, "learning_rate": 0.0002, "epoch": 1.2026143790849673, "step": 1840}, {"loss": 1.4343, "grad_norm": 0.6810969710350037, "learning_rate": 0.0002, "epoch": 1.2091503267973855, "step": 1850}, {"loss": 1.446, "grad_norm": 0.579741895198822, "learning_rate": 0.0002, "epoch": 1.215686274509804, "step": 1860}, {"loss": 1.4564, "grad_norm": 0.9925695657730103, "learning_rate": 0.0002, "epoch": 1.2222222222222223, "step": 1870}, {"loss": 1.5516, "grad_norm": 0.5919767618179321, "learning_rate": 0.0002, "epoch": 1.2287581699346406, "step": 1880}, {"loss": 1.5015, "grad_norm": 0.7377090454101562, "learning_rate": 0.0002, "epoch": 1.2352941176470589, "step": 1890}, {"loss": 1.4756, "grad_norm": 0.5753688812255859, "learning_rate": 0.0002, "epoch": 1.2418300653594772, "step": 1900}, {"loss": 1.3543, "grad_norm": 0.6362486481666565, "learning_rate": 0.0002, "epoch": 1.2483660130718954, "step": 1910}, {"loss": 1.4153, "grad_norm": 0.5747467875480652, "learning_rate": 0.0002, "epoch": 1.2549019607843137, "step": 1920}, {"loss": 1.5082, "grad_norm": 0.6831939220428467, "learning_rate": 0.0002, "epoch": 1.261437908496732, "step": 1930}, {"loss": 1.3509, "grad_norm": 0.6414040327072144, "learning_rate": 0.0002, "epoch": 1.2679738562091503, "step": 1940}, {"loss": 1.5099, "grad_norm": 0.5613330006599426, "learning_rate": 0.0002, "epoch": 1.2745098039215685, "step": 1950}, {"loss": 1.377, "grad_norm": 0.5838454961776733, "learning_rate": 0.0002, "epoch": 1.2810457516339868, "step": 1960}, {"loss": 1.3548, "grad_norm": 0.5367192029953003, "learning_rate": 0.0002, "epoch": 1.287581699346405, "step": 1970}, {"loss": 1.4602, "grad_norm": 0.5829346776008606, "learning_rate": 0.0002, "epoch": 1.2941176470588236, "step": 1980}, {"loss": 1.3821, "grad_norm": 0.756534218788147, "learning_rate": 0.0002, "epoch": 1.3006535947712419, "step": 1990}, {"loss": 1.389, "grad_norm": 0.48002561926841736, "learning_rate": 0.0002, "epoch": 1.3071895424836601, "step": 2000}, {"loss": 1.256, "grad_norm": 0.5461082458496094, "learning_rate": 0.0002, "epoch": 1.3137254901960784, "step": 2010}, {"loss": 1.6257, "grad_norm": 0.570399284362793, "learning_rate": 0.0002, "epoch": 1.3202614379084967, "step": 2020}, {"loss": 1.4356, "grad_norm": 0.5130975842475891, "learning_rate": 0.0002, "epoch": 1.326797385620915, "step": 2030}, {"loss": 1.3552, "grad_norm": 0.6290071606636047, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 2040}, {"loss": 1.3873, "grad_norm": 0.6165726184844971, "learning_rate": 0.0002, "epoch": 1.3398692810457518, "step": 2050}, {"loss": 1.4376, "grad_norm": 0.5302083492279053, "learning_rate": 0.0002, "epoch": 1.34640522875817, "step": 2060}, {"loss": 1.4722, "grad_norm": 0.6531406044960022, "learning_rate": 0.0002, "epoch": 1.3529411764705883, "step": 2070}, {"loss": 1.3632, "grad_norm": 0.5981236100196838, "learning_rate": 0.0002, "epoch": 1.3594771241830066, "step": 2080}, {"loss": 1.4846, "grad_norm": 0.8534150123596191, "learning_rate": 0.0002, "epoch": 1.3660130718954249, "step": 2090}, {"loss": 1.3249, "grad_norm": 0.695918083190918, "learning_rate": 0.0002, "epoch": 1.3725490196078431, "step": 2100}, {"loss": 1.4989, "grad_norm": 0.5830431580543518, "learning_rate": 0.0002, "epoch": 1.3790849673202614, "step": 2110}, {"loss": 1.5009, "grad_norm": 0.5641306638717651, "learning_rate": 0.0002, "epoch": 1.3856209150326797, "step": 2120}, {"loss": 1.3985, "grad_norm": 0.6354436874389648, "learning_rate": 0.0002, "epoch": 1.392156862745098, "step": 2130}, {"loss": 1.2737, "grad_norm": 0.5707540512084961, "learning_rate": 0.0002, "epoch": 1.3986928104575163, "step": 2140}, {"loss": 1.3815, "grad_norm": 0.7308434844017029, "learning_rate": 0.0002, "epoch": 1.4052287581699345, "step": 2150}, {"loss": 1.3993, "grad_norm": 0.5879750847816467, "learning_rate": 0.0002, "epoch": 1.4117647058823528, "step": 2160}, {"loss": 1.3729, "grad_norm": 0.627909243106842, "learning_rate": 0.0002, "epoch": 1.4183006535947713, "step": 2170}, {"loss": 1.3391, "grad_norm": 0.5228193998336792, "learning_rate": 0.0002, "epoch": 1.4248366013071896, "step": 2180}, {"loss": 1.457, "grad_norm": 0.6162880659103394, "learning_rate": 0.0002, "epoch": 1.4313725490196079, "step": 2190}, {"loss": 1.4052, "grad_norm": 0.751610517501831, "learning_rate": 0.0002, "epoch": 1.4379084967320261, "step": 2200}, {"loss": 1.4105, "grad_norm": 0.5623487234115601, "learning_rate": 0.0002, "epoch": 1.4444444444444444, "step": 2210}, {"loss": 1.3795, "grad_norm": 0.5293187499046326, "learning_rate": 0.0002, "epoch": 1.4509803921568627, "step": 2220}, {"loss": 1.4247, "grad_norm": 0.5903629660606384, "learning_rate": 0.0002, "epoch": 1.457516339869281, "step": 2230}, {"loss": 1.6167, "grad_norm": 0.6084659099578857, "learning_rate": 0.0002, "epoch": 1.4640522875816995, "step": 2240}, {"loss": 1.319, "grad_norm": 0.5289803147315979, "learning_rate": 0.0002, "epoch": 1.4705882352941178, "step": 2250}, {"loss": 1.3106, "grad_norm": 0.49499568343162537, "learning_rate": 0.0002, "epoch": 1.477124183006536, "step": 2260}, {"loss": 1.3586, "grad_norm": 0.7774190306663513, "learning_rate": 0.0002, "epoch": 1.4836601307189543, "step": 2270}, {"loss": 1.3075, "grad_norm": 0.5932538509368896, "learning_rate": 0.0002, "epoch": 1.4901960784313726, "step": 2280}, {"loss": 1.3241, "grad_norm": 0.6009492874145508, "learning_rate": 0.0002, "epoch": 1.4967320261437909, "step": 2290}, {"loss": 1.3728, "grad_norm": 0.5559343099594116, "learning_rate": 0.0002, "epoch": 1.5032679738562091, "step": 2300}, {"loss": 1.2379, "grad_norm": 0.5956196188926697, "learning_rate": 0.0002, "epoch": 1.5098039215686274, "step": 2310}, {"loss": 1.5292, "grad_norm": 0.5624083876609802, "learning_rate": 0.0002, "epoch": 1.5163398692810457, "step": 2320}, {"loss": 1.4779, "grad_norm": 0.7195250391960144, "learning_rate": 0.0002, "epoch": 1.522875816993464, "step": 2330}, {"loss": 1.2938, "grad_norm": 0.6010490655899048, "learning_rate": 0.0002, "epoch": 1.5294117647058822, "step": 2340}, {"loss": 1.4121, "grad_norm": 0.664929211139679, "learning_rate": 0.0002, "epoch": 1.5359477124183005, "step": 2350}, {"loss": 1.4362, "grad_norm": 0.5158776640892029, "learning_rate": 0.0002, "epoch": 1.5424836601307188, "step": 2360}, {"loss": 1.2157, "grad_norm": 0.5147154927253723, "learning_rate": 0.0002, "epoch": 1.5490196078431373, "step": 2370}, {"loss": 1.2643, "grad_norm": 0.6507977843284607, "learning_rate": 0.0002, "epoch": 1.5555555555555556, "step": 2380}, {"loss": 1.2786, "grad_norm": 0.5193192362785339, "learning_rate": 0.0002, "epoch": 1.5620915032679739, "step": 2390}, {"loss": 1.3209, "grad_norm": 0.5982314944267273, "learning_rate": 0.0002, "epoch": 1.5686274509803921, "step": 2400}, {"loss": 1.3585, "grad_norm": 0.49106258153915405, "learning_rate": 0.0002, "epoch": 1.5751633986928104, "step": 2410}, {"loss": 1.3618, "grad_norm": 0.6459611654281616, "learning_rate": 0.0002, "epoch": 1.581699346405229, "step": 2420}, {"loss": 1.3305, "grad_norm": 0.7038363218307495, "learning_rate": 0.0002, "epoch": 1.5882352941176472, "step": 2430}, {"loss": 1.3198, "grad_norm": 0.5245680212974548, "learning_rate": 0.0002, "epoch": 1.5947712418300655, "step": 2440}, {"loss": 1.4756, "grad_norm": 0.6562076210975647, "learning_rate": 0.0002, "epoch": 1.6013071895424837, "step": 2450}, {"loss": 1.5635, "grad_norm": 0.6491968035697937, "learning_rate": 0.0002, "epoch": 1.607843137254902, "step": 2460}, {"loss": 1.3657, "grad_norm": 0.604034960269928, "learning_rate": 0.0002, "epoch": 1.6143790849673203, "step": 2470}, {"loss": 1.2693, "grad_norm": 0.5759671330451965, "learning_rate": 0.0002, "epoch": 1.6209150326797386, "step": 2480}, {"loss": 1.4136, "grad_norm": 0.6157698631286621, "learning_rate": 0.0002, "epoch": 1.6274509803921569, "step": 2490}, {"loss": 1.3929, "grad_norm": 0.6513794660568237, "learning_rate": 0.0002, "epoch": 1.6339869281045751, "step": 2500}, {"loss": 1.4283, "grad_norm": 0.71990966796875, "learning_rate": 0.0002, "epoch": 1.6405228758169934, "step": 2510}, {"loss": 1.4356, "grad_norm": 0.7316617369651794, "learning_rate": 0.0002, "epoch": 1.6470588235294117, "step": 2520}, {"loss": 1.3119, "grad_norm": 0.5475177764892578, "learning_rate": 0.0002, "epoch": 1.65359477124183, "step": 2530}, {"loss": 1.2998, "grad_norm": 0.4911293089389801, "learning_rate": 0.0002, "epoch": 1.6601307189542482, "step": 2540}, {"loss": 1.4198, "grad_norm": 0.6122882962226868, "learning_rate": 0.0002, "epoch": 1.6666666666666665, "step": 2550}, {"loss": 1.3099, "grad_norm": 0.5735281705856323, "learning_rate": 0.0002, "epoch": 1.673202614379085, "step": 2560}, {"loss": 1.2205, "grad_norm": 0.5046352744102478, "learning_rate": 0.0002, "epoch": 1.6797385620915033, "step": 2570}, {"loss": 1.3191, "grad_norm": 0.6043242812156677, "learning_rate": 0.0002, "epoch": 1.6862745098039216, "step": 2580}, {"loss": 1.3079, "grad_norm": 0.5397698283195496, "learning_rate": 0.0002, "epoch": 1.6928104575163399, "step": 2590}, {"loss": 1.4916, "grad_norm": 0.8066475987434387, "learning_rate": 0.0002, "epoch": 1.6993464052287581, "step": 2600}, {"loss": 1.3703, "grad_norm": 0.52901691198349, "learning_rate": 0.0002, "epoch": 1.7058823529411766, "step": 2610}, {"loss": 1.409, "grad_norm": 0.7588503956794739, "learning_rate": 0.0002, "epoch": 1.712418300653595, "step": 2620}, {"loss": 1.3806, "grad_norm": 0.6012966632843018, "learning_rate": 0.0002, "epoch": 1.7189542483660132, "step": 2630}, {"loss": 1.2583, "grad_norm": 0.5927302837371826, "learning_rate": 0.0002, "epoch": 1.7254901960784315, "step": 2640}, {"loss": 1.4523, "grad_norm": 0.5086990594863892, "learning_rate": 0.0002, "epoch": 1.7320261437908497, "step": 2650}, {"loss": 1.5452, "grad_norm": 0.6000628471374512, "learning_rate": 0.0002, "epoch": 1.738562091503268, "step": 2660}, {"loss": 1.3269, "grad_norm": 0.6560431718826294, "learning_rate": 0.0002, "epoch": 1.7450980392156863, "step": 2670}, {"loss": 1.3982, "grad_norm": 0.5738165378570557, "learning_rate": 0.0002, "epoch": 1.7516339869281046, "step": 2680}, {"loss": 1.3766, "grad_norm": 0.5576106905937195, "learning_rate": 0.0002, "epoch": 1.7581699346405228, "step": 2690}, {"loss": 1.3277, "grad_norm": 0.7298802137374878, "learning_rate": 0.0002, "epoch": 1.7647058823529411, "step": 2700}, {"loss": 1.2618, "grad_norm": 0.5751826167106628, "learning_rate": 0.0002, "epoch": 1.7712418300653594, "step": 2710}, {"loss": 1.35, "grad_norm": 0.6069957613945007, "learning_rate": 0.0002, "epoch": 1.7777777777777777, "step": 2720}, {"loss": 1.3492, "grad_norm": 0.7513017654418945, "learning_rate": 0.0002, "epoch": 1.784313725490196, "step": 2730}, {"loss": 1.2979, "grad_norm": 0.6058869957923889, "learning_rate": 0.0002, "epoch": 1.7908496732026142, "step": 2740}, {"loss": 1.299, "grad_norm": 0.6805883049964905, "learning_rate": 0.0002, "epoch": 1.7973856209150327, "step": 2750}, {"loss": 1.4062, "grad_norm": 0.6864324808120728, "learning_rate": 0.0002, "epoch": 1.803921568627451, "step": 2760}, {"loss": 1.355, "grad_norm": 0.6261002421379089, "learning_rate": 0.0002, "epoch": 1.8104575163398693, "step": 2770}, {"loss": 1.5145, "grad_norm": 0.532684862613678, "learning_rate": 0.0002, "epoch": 1.8169934640522876, "step": 2780}, {"loss": 1.3248, "grad_norm": 0.6209020018577576, "learning_rate": 0.0002, "epoch": 1.8235294117647058, "step": 2790}, {"loss": 1.3908, "grad_norm": 0.67111736536026, "learning_rate": 0.0002, "epoch": 1.8300653594771243, "step": 2800}, {"loss": 1.5088, "grad_norm": 0.700467586517334, "learning_rate": 0.0002, "epoch": 1.8366013071895426, "step": 2810}, {"loss": 1.348, "grad_norm": 0.6968029141426086, "learning_rate": 0.0002, "epoch": 1.843137254901961, "step": 2820}, {"loss": 1.3943, "grad_norm": 0.6405863761901855, "learning_rate": 0.0002, "epoch": 1.8496732026143792, "step": 2830}, {"loss": 1.4035, "grad_norm": 0.5192584991455078, "learning_rate": 0.0002, "epoch": 1.8562091503267975, "step": 2840}, {"loss": 1.2745, "grad_norm": 0.4888569414615631, "learning_rate": 0.0002, "epoch": 1.8627450980392157, "step": 2850}, {"loss": 1.4324, "grad_norm": 0.7625455856323242, "learning_rate": 0.0002, "epoch": 1.869281045751634, "step": 2860}, {"loss": 1.4989, "grad_norm": 0.9162808656692505, "learning_rate": 0.0002, "epoch": 1.8758169934640523, "step": 2870}, {"loss": 1.3978, "grad_norm": 0.5472783446311951, "learning_rate": 0.0002, "epoch": 1.8823529411764706, "step": 2880}, {"loss": 1.3026, "grad_norm": 0.5221137404441833, "learning_rate": 0.0002, "epoch": 1.8888888888888888, "step": 2890}, {"loss": 1.33, "grad_norm": 0.49258849024772644, "learning_rate": 0.0002, "epoch": 1.8954248366013071, "step": 2900}, {"loss": 1.3503, "grad_norm": 0.5260750651359558, "learning_rate": 0.0002, "epoch": 1.9019607843137254, "step": 2910}, {"loss": 1.3381, "grad_norm": 0.6583314538002014, "learning_rate": 0.0002, "epoch": 1.9084967320261437, "step": 2920}, {"loss": 1.356, "grad_norm": 0.5728915929794312, "learning_rate": 0.0002, "epoch": 1.915032679738562, "step": 2930}, {"loss": 1.3993, "grad_norm": 0.7661453485488892, "learning_rate": 0.0002, "epoch": 1.9215686274509802, "step": 2940}, {"loss": 1.428, "grad_norm": 0.7193911075592041, "learning_rate": 0.0002, "epoch": 1.9281045751633987, "step": 2950}, {"loss": 1.287, "grad_norm": 0.5007768869400024, "learning_rate": 0.0002, "epoch": 1.934640522875817, "step": 2960}, {"loss": 1.372, "grad_norm": 0.626681923866272, "learning_rate": 0.0002, "epoch": 1.9411764705882353, "step": 2970}, {"loss": 1.375, "grad_norm": 0.8692840933799744, "learning_rate": 0.0002, "epoch": 1.9477124183006536, "step": 2980}, {"loss": 1.3292, "grad_norm": 0.6388291120529175, "learning_rate": 0.0002, "epoch": 1.954248366013072, "step": 2990}, {"loss": 1.4593, "grad_norm": 0.7710477113723755, "learning_rate": 0.0002, "epoch": 1.9607843137254903, "step": 3000}, {"loss": 1.5228, "grad_norm": 0.641704261302948, "learning_rate": 0.0002, "epoch": 1.9673202614379086, "step": 3010}, {"loss": 1.3246, "grad_norm": 0.621148943901062, "learning_rate": 0.0002, "epoch": 1.973856209150327, "step": 3020}, {"loss": 1.3017, "grad_norm": 0.5119547247886658, "learning_rate": 0.0002, "epoch": 1.9803921568627452, "step": 3030}, {"loss": 1.4923, "grad_norm": 0.8104137778282166, "learning_rate": 0.0002, "epoch": 1.9869281045751634, "step": 3040}, {"loss": 1.3331, "grad_norm": 0.5856240391731262, "learning_rate": 0.0002, "epoch": 1.9934640522875817, "step": 3050}, {"loss": 1.4346, "grad_norm": 0.5263566374778748, "learning_rate": 0.0002, "epoch": 2.0, "step": 3060}, {"eval_loss": 1.4276371002197266, "eval_runtime": 30.5759, "eval_samples_per_second": 14.26, "eval_steps_per_second": 1.799, "epoch": 2.0, "step": 3060}, {"loss": 1.1636, "grad_norm": 0.5143898725509644, "learning_rate": 0.0002, "epoch": 2.0065359477124183, "step": 3070}, {"loss": 1.3335, "grad_norm": 0.5749367475509644, "learning_rate": 0.0002, "epoch": 2.0130718954248366, "step": 3080}, {"loss": 1.2784, "grad_norm": 0.5784284472465515, "learning_rate": 0.0002, "epoch": 2.019607843137255, "step": 3090}, {"loss": 1.2463, "grad_norm": 0.5933429598808289, "learning_rate": 0.0002, "epoch": 2.026143790849673, "step": 3100}, {"loss": 1.2984, "grad_norm": 0.6748974919319153, "learning_rate": 0.0002, "epoch": 2.0326797385620914, "step": 3110}, {"loss": 1.2307, "grad_norm": 0.626399576663971, "learning_rate": 0.0002, "epoch": 2.0392156862745097, "step": 3120}, {"loss": 1.299, "grad_norm": 0.6173238754272461, "learning_rate": 0.0002, "epoch": 2.045751633986928, "step": 3130}, {"loss": 1.4144, "grad_norm": 0.807790219783783, "learning_rate": 0.0002, "epoch": 2.052287581699346, "step": 3140}, {"loss": 1.1953, "grad_norm": 0.6222215890884399, "learning_rate": 0.0002, "epoch": 2.0588235294117645, "step": 3150}, {"loss": 1.4059, "grad_norm": 0.5859580636024475, "learning_rate": 0.0002, "epoch": 2.065359477124183, "step": 3160}, {"loss": 1.3607, "grad_norm": 0.581304132938385, "learning_rate": 0.0002, "epoch": 2.0718954248366015, "step": 3170}, {"loss": 1.1212, "grad_norm": 0.9814971089363098, "learning_rate": 0.0002, "epoch": 2.0784313725490198, "step": 3180}, {"loss": 1.1962, "grad_norm": 0.6491848230361938, "learning_rate": 0.0002, "epoch": 2.084967320261438, "step": 3190}, {"loss": 1.3711, "grad_norm": 0.613680362701416, "learning_rate": 0.0002, "epoch": 2.0915032679738563, "step": 3200}, {"loss": 1.2994, "grad_norm": 0.7318086624145508, "learning_rate": 0.0002, "epoch": 2.0980392156862746, "step": 3210}, {"loss": 1.2502, "grad_norm": 0.6025661826133728, "learning_rate": 0.0002, "epoch": 2.104575163398693, "step": 3220}, {"loss": 1.1374, "grad_norm": 0.6744484305381775, "learning_rate": 0.0002, "epoch": 2.111111111111111, "step": 3230}, {"loss": 1.3273, "grad_norm": 0.6062554121017456, "learning_rate": 0.0002, "epoch": 2.1176470588235294, "step": 3240}, {"loss": 1.3404, "grad_norm": 0.6801803112030029, "learning_rate": 0.0002, "epoch": 2.1241830065359477, "step": 3250}, {"loss": 1.4084, "grad_norm": 0.5218925476074219, "learning_rate": 0.0002, "epoch": 2.130718954248366, "step": 3260}, {"loss": 1.2867, "grad_norm": 0.7494263648986816, "learning_rate": 0.0002, "epoch": 2.1372549019607843, "step": 3270}, {"loss": 1.3059, "grad_norm": 0.7858565449714661, "learning_rate": 0.0002, "epoch": 2.1437908496732025, "step": 3280}, {"loss": 1.3214, "grad_norm": 0.6836692690849304, "learning_rate": 0.0002, "epoch": 2.150326797385621, "step": 3290}, {"loss": 1.1605, "grad_norm": 0.619848370552063, "learning_rate": 0.0002, "epoch": 2.156862745098039, "step": 3300}, {"loss": 1.3095, "grad_norm": 0.5761294364929199, "learning_rate": 0.0002, "epoch": 2.1633986928104574, "step": 3310}, {"loss": 1.2883, "grad_norm": 0.4713786542415619, "learning_rate": 0.0002, "epoch": 2.1699346405228757, "step": 3320}, {"loss": 1.3817, "grad_norm": 0.7613773345947266, "learning_rate": 0.0002, "epoch": 2.176470588235294, "step": 3330}, {"loss": 1.2354, "grad_norm": 0.6642718315124512, "learning_rate": 0.0002, "epoch": 2.183006535947712, "step": 3340}, {"loss": 1.2048, "grad_norm": 0.7162188291549683, "learning_rate": 0.0002, "epoch": 2.189542483660131, "step": 3350}, {"loss": 1.3886, "grad_norm": 0.6916783452033997, "learning_rate": 0.0002, "epoch": 2.196078431372549, "step": 3360}, {"loss": 1.3788, "grad_norm": 0.7205567955970764, "learning_rate": 0.0002, "epoch": 2.2026143790849675, "step": 3370}, {"loss": 1.2528, "grad_norm": 0.6038199067115784, "learning_rate": 0.0002, "epoch": 2.2091503267973858, "step": 3380}, {"loss": 1.2079, "grad_norm": 0.6284233927726746, "learning_rate": 0.0002, "epoch": 2.215686274509804, "step": 3390}, {"loss": 1.3057, "grad_norm": 0.7450672388076782, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3400}, {"loss": 1.3034, "grad_norm": 0.7755052447319031, "learning_rate": 0.0002, "epoch": 2.2287581699346406, "step": 3410}, {"loss": 1.2953, "grad_norm": 0.9066099524497986, "learning_rate": 0.0002, "epoch": 2.235294117647059, "step": 3420}, {"loss": 1.3072, "grad_norm": 0.8578207492828369, "learning_rate": 0.0002, "epoch": 2.241830065359477, "step": 3430}, {"loss": 1.3278, "grad_norm": 0.5900213718414307, "learning_rate": 0.0002, "epoch": 2.2483660130718954, "step": 3440}, {"loss": 1.3645, "grad_norm": 0.7821717262268066, "learning_rate": 0.0002, "epoch": 2.2549019607843137, "step": 3450}, {"loss": 1.183, "grad_norm": 0.6263150572776794, "learning_rate": 0.0002, "epoch": 2.261437908496732, "step": 3460}, {"loss": 1.178, "grad_norm": 0.591799259185791, "learning_rate": 0.0002, "epoch": 2.2679738562091503, "step": 3470}, {"loss": 1.2198, "grad_norm": 0.5999799966812134, "learning_rate": 0.0002, "epoch": 2.2745098039215685, "step": 3480}, {"loss": 1.2724, "grad_norm": 0.6227319240570068, "learning_rate": 0.0002, "epoch": 2.281045751633987, "step": 3490}, {"loss": 1.3865, "grad_norm": 0.719412624835968, "learning_rate": 0.0002, "epoch": 2.287581699346405, "step": 3500}, {"loss": 1.3275, "grad_norm": 1.0361769199371338, "learning_rate": 0.0002, "epoch": 2.2941176470588234, "step": 3510}, {"loss": 1.4834, "grad_norm": 0.5506668090820312, "learning_rate": 0.0002, "epoch": 2.3006535947712417, "step": 3520}, {"loss": 1.2273, "grad_norm": 0.6886829733848572, "learning_rate": 0.0002, "epoch": 2.30718954248366, "step": 3530}, {"loss": 1.2296, "grad_norm": 0.6226346492767334, "learning_rate": 0.0002, "epoch": 2.313725490196078, "step": 3540}, {"loss": 1.3087, "grad_norm": 0.8109908103942871, "learning_rate": 0.0002, "epoch": 2.3202614379084965, "step": 3550}, {"loss": 1.3311, "grad_norm": 0.8505511283874512, "learning_rate": 0.0002, "epoch": 2.326797385620915, "step": 3560}, {"loss": 1.2526, "grad_norm": 0.5763760209083557, "learning_rate": 0.0002, "epoch": 2.3333333333333335, "step": 3570}, {"loss": 1.4135, "grad_norm": 0.6460059881210327, "learning_rate": 0.0002, "epoch": 2.3398692810457518, "step": 3580}, {"loss": 1.2701, "grad_norm": 0.7175343036651611, "learning_rate": 0.0002, "epoch": 2.34640522875817, "step": 3590}, {"loss": 1.2645, "grad_norm": 0.6012630462646484, "learning_rate": 0.0002, "epoch": 2.3529411764705883, "step": 3600}, {"loss": 1.3214, "grad_norm": 0.6513685584068298, "learning_rate": 0.0002, "epoch": 2.3594771241830066, "step": 3610}, {"loss": 1.3271, "grad_norm": 0.7465183734893799, "learning_rate": 0.0002, "epoch": 2.366013071895425, "step": 3620}, {"loss": 1.3671, "grad_norm": 0.6413124203681946, "learning_rate": 0.0002, "epoch": 2.372549019607843, "step": 3630}, {"loss": 1.4026, "grad_norm": 0.7209562063217163, "learning_rate": 0.0002, "epoch": 2.3790849673202614, "step": 3640}, {"loss": 1.1616, "grad_norm": 0.6427558660507202, "learning_rate": 0.0002, "epoch": 2.3856209150326797, "step": 3650}, {"loss": 1.313, "grad_norm": 0.593958854675293, "learning_rate": 0.0002, "epoch": 2.392156862745098, "step": 3660}, {"loss": 1.2802, "grad_norm": 0.5944608449935913, "learning_rate": 0.0002, "epoch": 2.3986928104575163, "step": 3670}, {"loss": 1.3542, "grad_norm": 0.6606248617172241, "learning_rate": 0.0002, "epoch": 2.4052287581699345, "step": 3680}, {"loss": 1.2977, "grad_norm": 0.5632851719856262, "learning_rate": 0.0002, "epoch": 2.411764705882353, "step": 3690}, {"loss": 1.2032, "grad_norm": 0.4976513385772705, "learning_rate": 0.0002, "epoch": 2.418300653594771, "step": 3700}, {"loss": 1.1404, "grad_norm": 0.6318528056144714, "learning_rate": 0.0002, "epoch": 2.4248366013071894, "step": 3710}, {"loss": 1.1705, "grad_norm": 0.6306707859039307, "learning_rate": 0.0002, "epoch": 2.431372549019608, "step": 3720}, {"loss": 1.3524, "grad_norm": 0.6362553238868713, "learning_rate": 0.0002, "epoch": 2.4379084967320264, "step": 3730}, {"loss": 1.2345, "grad_norm": 0.634368896484375, "learning_rate": 0.0002, "epoch": 2.4444444444444446, "step": 3740}, {"loss": 1.2515, "grad_norm": 0.6623591184616089, "learning_rate": 0.0002, "epoch": 2.450980392156863, "step": 3750}, {"loss": 1.3246, "grad_norm": 0.6150440573692322, "learning_rate": 0.0002, "epoch": 2.457516339869281, "step": 3760}, {"loss": 1.2666, "grad_norm": 0.588935911655426, "learning_rate": 0.0002, "epoch": 2.4640522875816995, "step": 3770}, {"loss": 1.3918, "grad_norm": 0.7388206124305725, "learning_rate": 0.0002, "epoch": 2.4705882352941178, "step": 3780}, {"loss": 1.2512, "grad_norm": 0.621825098991394, "learning_rate": 0.0002, "epoch": 2.477124183006536, "step": 3790}, {"loss": 1.359, "grad_norm": 0.7691677212715149, "learning_rate": 0.0002, "epoch": 2.4836601307189543, "step": 3800}, {"loss": 1.3399, "grad_norm": 1.1661969423294067, "learning_rate": 0.0002, "epoch": 2.4901960784313726, "step": 3810}, {"loss": 1.461, "grad_norm": 0.6837884187698364, "learning_rate": 0.0002, "epoch": 2.496732026143791, "step": 3820}, {"loss": 1.2823, "grad_norm": 0.6978904008865356, "learning_rate": 0.0002, "epoch": 2.503267973856209, "step": 3830}, {"loss": 1.3688, "grad_norm": 0.6121411323547363, "learning_rate": 0.0002, "epoch": 2.5098039215686274, "step": 3840}, {"loss": 1.2587, "grad_norm": 0.7813326120376587, "learning_rate": 0.0002, "epoch": 2.5163398692810457, "step": 3850}, {"loss": 1.1543, "grad_norm": 0.5390260219573975, "learning_rate": 0.0002, "epoch": 2.522875816993464, "step": 3860}, {"loss": 1.2032, "grad_norm": 0.8283252716064453, "learning_rate": 0.0002, "epoch": 2.5294117647058822, "step": 3870}, {"loss": 1.3112, "grad_norm": 0.8527186512947083, "learning_rate": 0.0002, "epoch": 2.5359477124183005, "step": 3880}, {"loss": 1.3469, "grad_norm": 0.8405382633209229, "learning_rate": 0.0002, "epoch": 2.542483660130719, "step": 3890}, {"loss": 1.1801, "grad_norm": 0.5650738477706909, "learning_rate": 0.0002, "epoch": 2.549019607843137, "step": 3900}, {"loss": 1.2917, "grad_norm": 0.620121955871582, "learning_rate": 0.0002, "epoch": 2.5555555555555554, "step": 3910}, {"loss": 1.2524, "grad_norm": 0.5983527898788452, "learning_rate": 0.0002, "epoch": 2.5620915032679736, "step": 3920}, {"loss": 1.4408, "grad_norm": 0.686623215675354, "learning_rate": 0.0002, "epoch": 2.568627450980392, "step": 3930}, {"loss": 1.186, "grad_norm": 0.6805831789970398, "learning_rate": 0.0002, "epoch": 2.57516339869281, "step": 3940}, {"loss": 1.367, "grad_norm": 0.6994825601577759, "learning_rate": 0.0002, "epoch": 2.581699346405229, "step": 3950}, {"loss": 1.3446, "grad_norm": 0.728549599647522, "learning_rate": 0.0002, "epoch": 2.588235294117647, "step": 3960}, {"loss": 1.4039, "grad_norm": 0.775236964225769, "learning_rate": 0.0002, "epoch": 2.5947712418300655, "step": 3970}, {"loss": 1.2742, "grad_norm": 0.5057447552680969, "learning_rate": 0.0002, "epoch": 2.6013071895424837, "step": 3980}, {"loss": 1.2764, "grad_norm": 0.6564450263977051, "learning_rate": 0.0002, "epoch": 2.607843137254902, "step": 3990}, {"loss": 1.3269, "grad_norm": 0.5342249870300293, "learning_rate": 0.0002, "epoch": 2.6143790849673203, "step": 4000}, {"loss": 1.3102, "grad_norm": 0.5508961081504822, "learning_rate": 0.0002, "epoch": 2.6209150326797386, "step": 4010}, {"loss": 1.3636, "grad_norm": 0.5716235637664795, "learning_rate": 0.0002, "epoch": 2.627450980392157, "step": 4020}, {"loss": 1.3465, "grad_norm": 0.8049232363700867, "learning_rate": 0.0002, "epoch": 2.633986928104575, "step": 4030}, {"loss": 1.2342, "grad_norm": 0.5574354529380798, "learning_rate": 0.0002, "epoch": 2.6405228758169934, "step": 4040}, {"loss": 1.2419, "grad_norm": 0.6302093863487244, "learning_rate": 0.0002, "epoch": 2.6470588235294117, "step": 4050}, {"loss": 1.2565, "grad_norm": 1.1868736743927002, "learning_rate": 0.0002, "epoch": 2.65359477124183, "step": 4060}, {"loss": 1.1382, "grad_norm": 0.6738120317459106, "learning_rate": 0.0002, "epoch": 2.6601307189542482, "step": 4070}, {"loss": 1.2456, "grad_norm": 0.6614423990249634, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 4080}, {"loss": 1.2958, "grad_norm": 0.7297604084014893, "learning_rate": 0.0002, "epoch": 2.6732026143790852, "step": 4090}, {"loss": 1.1596, "grad_norm": 0.9421682357788086, "learning_rate": 0.0002, "epoch": 2.6797385620915035, "step": 4100}, {"loss": 1.3002, "grad_norm": 0.5286222696304321, "learning_rate": 0.0002, "epoch": 2.686274509803922, "step": 4110}, {"loss": 1.3936, "grad_norm": 0.6849271655082703, "learning_rate": 0.0002, "epoch": 2.69281045751634, "step": 4120}, {"loss": 1.2721, "grad_norm": 0.6811320185661316, "learning_rate": 0.0002, "epoch": 2.6993464052287583, "step": 4130}, {"loss": 1.2897, "grad_norm": 0.4968419373035431, "learning_rate": 0.0002, "epoch": 2.7058823529411766, "step": 4140}, {"loss": 1.3322, "grad_norm": 0.8074267506599426, "learning_rate": 0.0002, "epoch": 2.712418300653595, "step": 4150}, {"loss": 1.1759, "grad_norm": 0.6756376028060913, "learning_rate": 0.0002, "epoch": 2.718954248366013, "step": 4160}, {"loss": 1.2444, "grad_norm": 0.6921583414077759, "learning_rate": 0.0002, "epoch": 2.7254901960784315, "step": 4170}, {"loss": 1.3413, "grad_norm": 0.7049834132194519, "learning_rate": 0.0002, "epoch": 2.7320261437908497, "step": 4180}, {"loss": 1.1965, "grad_norm": 0.7011390328407288, "learning_rate": 0.0002, "epoch": 2.738562091503268, "step": 4190}, {"loss": 1.2364, "grad_norm": 0.6977843642234802, "learning_rate": 0.0002, "epoch": 2.7450980392156863, "step": 4200}, {"loss": 1.2533, "grad_norm": 0.6717000603675842, "learning_rate": 0.0002, "epoch": 2.7516339869281046, "step": 4210}, {"loss": 1.392, "grad_norm": 1.0223724842071533, "learning_rate": 0.0002, "epoch": 2.758169934640523, "step": 4220}, {"loss": 1.2451, "grad_norm": 0.6573330760002136, "learning_rate": 0.0002, "epoch": 2.764705882352941, "step": 4230}, {"loss": 1.4219, "grad_norm": 0.6684938073158264, "learning_rate": 0.0002, "epoch": 2.7712418300653594, "step": 4240}, {"loss": 1.2505, "grad_norm": 0.7426793575286865, "learning_rate": 0.0002, "epoch": 2.7777777777777777, "step": 4250}, {"loss": 1.2904, "grad_norm": 0.557826578617096, "learning_rate": 0.0002, "epoch": 2.784313725490196, "step": 4260}, {"loss": 1.3262, "grad_norm": 0.6669870018959045, "learning_rate": 0.0002, "epoch": 2.7908496732026142, "step": 4270}, {"loss": 1.2369, "grad_norm": 0.5349969267845154, "learning_rate": 0.0002, "epoch": 2.7973856209150325, "step": 4280}, {"loss": 1.3769, "grad_norm": 0.7262802124023438, "learning_rate": 0.0002, "epoch": 2.803921568627451, "step": 4290}, {"loss": 1.3373, "grad_norm": 0.768211841583252, "learning_rate": 0.0002, "epoch": 2.810457516339869, "step": 4300}, {"loss": 1.2444, "grad_norm": 0.5958252549171448, "learning_rate": 0.0002, "epoch": 2.8169934640522873, "step": 4310}, {"loss": 1.4113, "grad_norm": 0.8451310396194458, "learning_rate": 0.0002, "epoch": 2.8235294117647056, "step": 4320}, {"loss": 1.2454, "grad_norm": 0.6544435024261475, "learning_rate": 0.0002, "epoch": 2.8300653594771243, "step": 4330}, {"loss": 1.2777, "grad_norm": 0.6177433133125305, "learning_rate": 0.0002, "epoch": 2.8366013071895426, "step": 4340}, {"loss": 1.2562, "grad_norm": 0.6324988007545471, "learning_rate": 0.0002, "epoch": 2.843137254901961, "step": 4350}, {"loss": 1.4117, "grad_norm": 0.6884300708770752, "learning_rate": 0.0002, "epoch": 2.849673202614379, "step": 4360}, {"loss": 1.2391, "grad_norm": 0.8952897191047668, "learning_rate": 0.0002, "epoch": 2.8562091503267975, "step": 4370}, {"loss": 1.2814, "grad_norm": 1.0260103940963745, "learning_rate": 0.0002, "epoch": 2.8627450980392157, "step": 4380}, {"loss": 1.2893, "grad_norm": 0.9134647250175476, "learning_rate": 0.0002, "epoch": 2.869281045751634, "step": 4390}, {"loss": 1.171, "grad_norm": 0.5637717843055725, "learning_rate": 0.0002, "epoch": 2.8758169934640523, "step": 4400}, {"loss": 1.3422, "grad_norm": 0.7530393004417419, "learning_rate": 0.0002, "epoch": 2.8823529411764706, "step": 4410}, {"loss": 1.29, "grad_norm": 0.7202680706977844, "learning_rate": 0.0002, "epoch": 2.888888888888889, "step": 4420}, {"loss": 1.2913, "grad_norm": 0.7177144885063171, "learning_rate": 0.0002, "epoch": 2.895424836601307, "step": 4430}, {"loss": 1.1922, "grad_norm": 0.5996816754341125, "learning_rate": 0.0002, "epoch": 2.9019607843137254, "step": 4440}, {"loss": 1.4816, "grad_norm": 0.6542447209358215, "learning_rate": 0.0002, "epoch": 2.9084967320261437, "step": 4450}, {"loss": 1.503, "grad_norm": 1.0753740072250366, "learning_rate": 0.0002, "epoch": 2.915032679738562, "step": 4460}, {"loss": 1.3193, "grad_norm": 0.6956136226654053, "learning_rate": 0.0002, "epoch": 2.9215686274509802, "step": 4470}, {"loss": 1.2486, "grad_norm": 0.7702530026435852, "learning_rate": 0.0002, "epoch": 2.928104575163399, "step": 4480}, {"loss": 1.3371, "grad_norm": 0.7763232588768005, "learning_rate": 0.0002, "epoch": 2.9346405228758172, "step": 4490}, {"loss": 1.1647, "grad_norm": 0.6393085718154907, "learning_rate": 0.0002, "epoch": 2.9411764705882355, "step": 4500}, {"loss": 1.211, "grad_norm": 0.987770676612854, "learning_rate": 0.0002, "epoch": 2.947712418300654, "step": 4510}, {"loss": 1.1529, "grad_norm": 0.5995016098022461, "learning_rate": 0.0002, "epoch": 2.954248366013072, "step": 4520}, {"loss": 1.2358, "grad_norm": 0.745650053024292, "learning_rate": 0.0002, "epoch": 2.9607843137254903, "step": 4530}, {"loss": 1.2115, "grad_norm": 0.7429282069206238, "learning_rate": 0.0002, "epoch": 2.9673202614379086, "step": 4540}, {"loss": 1.2262, "grad_norm": 0.5927486419677734, "learning_rate": 0.0002, "epoch": 2.973856209150327, "step": 4550}, {"loss": 1.3173, "grad_norm": 0.6775153875350952, "learning_rate": 0.0002, "epoch": 2.980392156862745, "step": 4560}, {"loss": 1.279, "grad_norm": 0.7128435373306274, "learning_rate": 0.0002, "epoch": 2.9869281045751634, "step": 4570}, {"loss": 1.2451, "grad_norm": 0.7470937967300415, "learning_rate": 0.0002, "epoch": 2.9934640522875817, "step": 4580}, {"loss": 1.2701, "grad_norm": 0.9295375943183899, "learning_rate": 0.0002, "epoch": 3.0, "step": 4590}, {"eval_loss": 1.4131312370300293, "eval_runtime": 31.8967, "eval_samples_per_second": 13.669, "eval_steps_per_second": 1.724, "epoch": 3.0, "step": 4590}, {"loss": 1.1283, "grad_norm": 0.6926420331001282, "learning_rate": 0.0002, "epoch": 3.0065359477124183, "step": 4600}, {"loss": 1.1537, "grad_norm": 0.6656355857849121, "learning_rate": 0.0002, "epoch": 3.0130718954248366, "step": 4610}, {"loss": 1.308, "grad_norm": 0.9901936650276184, "learning_rate": 0.0002, "epoch": 3.019607843137255, "step": 4620}, {"loss": 1.22, "grad_norm": 0.6713474988937378, "learning_rate": 0.0002, "epoch": 3.026143790849673, "step": 4630}, {"loss": 1.2249, "grad_norm": 0.6199324131011963, "learning_rate": 0.0002, "epoch": 3.0326797385620914, "step": 4640}, {"loss": 1.242, "grad_norm": 0.7180785536766052, "learning_rate": 0.0002, "epoch": 3.0392156862745097, "step": 4650}, {"loss": 1.1349, "grad_norm": 0.8256588578224182, "learning_rate": 0.0002, "epoch": 3.045751633986928, "step": 4660}, {"loss": 1.1431, "grad_norm": 0.6637389063835144, "learning_rate": 0.0002, "epoch": 3.052287581699346, "step": 4670}, {"loss": 1.1096, "grad_norm": 0.6980698108673096, "learning_rate": 0.0002, "epoch": 3.0588235294117645, "step": 4680}, {"loss": 1.196, "grad_norm": 0.8091534972190857, "learning_rate": 0.0002, "epoch": 3.065359477124183, "step": 4690}, {"loss": 1.1652, "grad_norm": 0.5715174078941345, "learning_rate": 0.0002, "epoch": 3.0718954248366015, "step": 4700}, {"loss": 1.1427, "grad_norm": 0.735639750957489, "learning_rate": 0.0002, "epoch": 3.0784313725490198, "step": 4710}, {"loss": 1.1522, "grad_norm": 0.7619708180427551, "learning_rate": 0.0002, "epoch": 3.084967320261438, "step": 4720}, {"loss": 1.0853, "grad_norm": 1.263566017150879, "learning_rate": 0.0002, "epoch": 3.0915032679738563, "step": 4730}, {"loss": 1.1348, "grad_norm": 0.6600871682167053, "learning_rate": 0.0002, "epoch": 3.0980392156862746, "step": 4740}, {"loss": 1.1766, "grad_norm": 0.717792809009552, "learning_rate": 0.0002, "epoch": 3.104575163398693, "step": 4750}, {"loss": 1.088, "grad_norm": 0.853714644908905, "learning_rate": 0.0002, "epoch": 3.111111111111111, "step": 4760}, {"loss": 1.2031, "grad_norm": 1.1004153490066528, "learning_rate": 0.0002, "epoch": 3.1176470588235294, "step": 4770}, {"loss": 1.3295, "grad_norm": 0.8566235899925232, "learning_rate": 0.0002, "epoch": 3.1241830065359477, "step": 4780}, {"loss": 1.2436, "grad_norm": 0.8315296173095703, "learning_rate": 0.0002, "epoch": 3.130718954248366, "step": 4790}, {"loss": 1.32, "grad_norm": 0.8020524978637695, "learning_rate": 0.0002, "epoch": 3.1372549019607843, "step": 4800}, {"loss": 1.1238, "grad_norm": 0.7564275860786438, "learning_rate": 0.0002, "epoch": 3.1437908496732025, "step": 4810}, {"loss": 1.1244, "grad_norm": 0.9077776670455933, "learning_rate": 0.0002, "epoch": 3.150326797385621, "step": 4820}, {"loss": 1.1399, "grad_norm": 0.6323099732398987, "learning_rate": 0.0002, "epoch": 3.156862745098039, "step": 4830}, {"loss": 1.1983, "grad_norm": 0.6625368595123291, "learning_rate": 0.0002, "epoch": 3.1633986928104574, "step": 4840}, {"loss": 1.066, "grad_norm": 0.8119261860847473, "learning_rate": 0.0002, "epoch": 3.1699346405228757, "step": 4850}, {"loss": 1.0224, "grad_norm": 0.6399450898170471, "learning_rate": 0.0002, "epoch": 3.176470588235294, "step": 4860}, {"loss": 1.2181, "grad_norm": 1.0659016370773315, "learning_rate": 0.0002, "epoch": 3.183006535947712, "step": 4870}, {"loss": 1.2914, "grad_norm": 0.8040369749069214, "learning_rate": 0.0002, "epoch": 3.189542483660131, "step": 4880}, {"loss": 1.1996, "grad_norm": 0.7784733176231384, "learning_rate": 0.0002, "epoch": 3.196078431372549, "step": 4890}, {"loss": 1.2051, "grad_norm": 0.9660294651985168, "learning_rate": 0.0002, "epoch": 3.2026143790849675, "step": 4900}, {"loss": 1.0419, "grad_norm": 1.0676977634429932, "learning_rate": 0.0002, "epoch": 3.2091503267973858, "step": 4910}, {"loss": 1.0083, "grad_norm": 0.5877565741539001, "learning_rate": 0.0002, "epoch": 3.215686274509804, "step": 4920}, {"loss": 1.1046, "grad_norm": 0.6164032816886902, "learning_rate": 0.0002, "epoch": 3.2222222222222223, "step": 4930}, {"loss": 1.1079, "grad_norm": 0.7627606987953186, "learning_rate": 0.0002, "epoch": 3.2287581699346406, "step": 4940}, {"loss": 1.2453, "grad_norm": 0.7442803978919983, "learning_rate": 0.0002, "epoch": 3.235294117647059, "step": 4950}, {"loss": 1.1087, "grad_norm": 0.7277812361717224, "learning_rate": 0.0002, "epoch": 3.241830065359477, "step": 4960}, {"loss": 1.2237, "grad_norm": 1.0301902294158936, "learning_rate": 0.0002, "epoch": 3.2483660130718954, "step": 4970}, {"loss": 1.1466, "grad_norm": 0.7798232436180115, "learning_rate": 0.0002, "epoch": 3.2549019607843137, "step": 4980}, {"loss": 1.2142, "grad_norm": 1.210265874862671, "learning_rate": 0.0002, "epoch": 3.261437908496732, "step": 4990}, {"loss": 1.1557, "grad_norm": 0.6677713990211487, "learning_rate": 0.0002, "epoch": 3.2679738562091503, "step": 5000}, {"loss": 1.3294, "grad_norm": 1.0524500608444214, "learning_rate": 0.0002, "epoch": 3.2745098039215685, "step": 5010}, {"loss": 1.1939, "grad_norm": 0.7091745734214783, "learning_rate": 0.0002, "epoch": 3.281045751633987, "step": 5020}, {"loss": 1.1891, "grad_norm": 0.8523224592208862, "learning_rate": 0.0002, "epoch": 3.287581699346405, "step": 5030}, {"loss": 1.1925, "grad_norm": 0.6120608448982239, "learning_rate": 0.0002, "epoch": 3.2941176470588234, "step": 5040}, {"loss": 1.0603, "grad_norm": 0.7437472939491272, "learning_rate": 0.0002, "epoch": 3.3006535947712417, "step": 5050}, {"loss": 1.1295, "grad_norm": 0.7611715197563171, "learning_rate": 0.0002, "epoch": 3.30718954248366, "step": 5060}, {"loss": 1.0531, "grad_norm": 0.7249704003334045, "learning_rate": 0.0002, "epoch": 3.313725490196078, "step": 5070}, {"loss": 1.2292, "grad_norm": 0.7316247820854187, "learning_rate": 0.0002, "epoch": 3.3202614379084965, "step": 5080}, {"loss": 1.1974, "grad_norm": 0.562412440776825, "learning_rate": 0.0002, "epoch": 3.326797385620915, "step": 5090}, {"loss": 1.0736, "grad_norm": 0.7052176594734192, "learning_rate": 0.0002, "epoch": 3.3333333333333335, "step": 5100}, {"loss": 1.122, "grad_norm": 0.7714211344718933, "learning_rate": 0.0002, "epoch": 3.3398692810457518, "step": 5110}, {"loss": 1.1684, "grad_norm": 1.0436055660247803, "learning_rate": 0.0002, "epoch": 3.34640522875817, "step": 5120}, {"loss": 1.0945, "grad_norm": 0.8867271542549133, "learning_rate": 0.0002, "epoch": 3.3529411764705883, "step": 5130}, {"loss": 1.159, "grad_norm": 0.8371267914772034, "learning_rate": 0.0002, "epoch": 3.3594771241830066, "step": 5140}, {"loss": 1.1073, "grad_norm": 0.7257837057113647, "learning_rate": 0.0002, "epoch": 3.366013071895425, "step": 5150}, {"loss": 1.1162, "grad_norm": 0.7102002501487732, "learning_rate": 0.0002, "epoch": 3.372549019607843, "step": 5160}, {"loss": 1.2056, "grad_norm": 0.7636350393295288, "learning_rate": 0.0002, "epoch": 3.3790849673202614, "step": 5170}, {"loss": 1.0708, "grad_norm": 0.6887359619140625, "learning_rate": 0.0002, "epoch": 3.3856209150326797, "step": 5180}, {"loss": 1.3807, "grad_norm": 0.8141424655914307, "learning_rate": 0.0002, "epoch": 3.392156862745098, "step": 5190}, {"loss": 1.1986, "grad_norm": 0.694423496723175, "learning_rate": 0.0002, "epoch": 3.3986928104575163, "step": 5200}, {"loss": 1.2945, "grad_norm": 0.914013683795929, "learning_rate": 0.0002, "epoch": 3.4052287581699345, "step": 5210}, {"loss": 1.1413, "grad_norm": 0.8503239750862122, "learning_rate": 0.0002, "epoch": 3.411764705882353, "step": 5220}, {"loss": 1.2696, "grad_norm": 0.6196836233139038, "learning_rate": 0.0002, "epoch": 3.418300653594771, "step": 5230}, {"loss": 1.2431, "grad_norm": 1.0760811567306519, "learning_rate": 0.0002, "epoch": 3.4248366013071894, "step": 5240}, {"loss": 1.1686, "grad_norm": 0.6524698138237, "learning_rate": 0.0002, "epoch": 3.431372549019608, "step": 5250}, {"loss": 1.2012, "grad_norm": 0.674467921257019, "learning_rate": 0.0002, "epoch": 3.4379084967320264, "step": 5260}, {"loss": 1.1015, "grad_norm": 0.7690372467041016, "learning_rate": 0.0002, "epoch": 3.4444444444444446, "step": 5270}, {"loss": 1.2511, "grad_norm": 0.8751813769340515, "learning_rate": 0.0002, "epoch": 3.450980392156863, "step": 5280}, {"loss": 1.1841, "grad_norm": 0.750407874584198, "learning_rate": 0.0002, "epoch": 3.457516339869281, "step": 5290}, {"loss": 1.0605, "grad_norm": 0.5991823077201843, "learning_rate": 0.0002, "epoch": 3.4640522875816995, "step": 5300}, {"loss": 1.2347, "grad_norm": 1.0164772272109985, "learning_rate": 0.0002, "epoch": 3.4705882352941178, "step": 5310}, {"loss": 1.2354, "grad_norm": 0.8704105019569397, "learning_rate": 0.0002, "epoch": 3.477124183006536, "step": 5320}, {"loss": 1.2169, "grad_norm": 0.709102213382721, "learning_rate": 0.0002, "epoch": 3.4836601307189543, "step": 5330}, {"loss": 1.2425, "grad_norm": 0.6273632049560547, "learning_rate": 0.0002, "epoch": 3.4901960784313726, "step": 5340}, {"loss": 1.1585, "grad_norm": 0.6807359457015991, "learning_rate": 0.0002, "epoch": 3.496732026143791, "step": 5350}, {"loss": 1.131, "grad_norm": 0.7085188627243042, "learning_rate": 0.0002, "epoch": 3.503267973856209, "step": 5360}, {"loss": 1.1159, "grad_norm": 0.6938307881355286, "learning_rate": 0.0002, "epoch": 3.5098039215686274, "step": 5370}, {"loss": 1.1397, "grad_norm": 0.8544146418571472, "learning_rate": 0.0002, "epoch": 3.5163398692810457, "step": 5380}, {"loss": 1.2181, "grad_norm": 0.7889642119407654, "learning_rate": 0.0002, "epoch": 3.522875816993464, "step": 5390}, {"loss": 1.1691, "grad_norm": 0.7858421206474304, "learning_rate": 0.0002, "epoch": 3.5294117647058822, "step": 5400}, {"loss": 1.2374, "grad_norm": 0.8547123074531555, "learning_rate": 0.0002, "epoch": 3.5359477124183005, "step": 5410}, {"loss": 1.196, "grad_norm": 0.8218181133270264, "learning_rate": 0.0002, "epoch": 3.542483660130719, "step": 5420}, {"loss": 1.1961, "grad_norm": 1.153623342514038, "learning_rate": 0.0002, "epoch": 3.549019607843137, "step": 5430}, {"loss": 1.156, "grad_norm": 1.1321099996566772, "learning_rate": 0.0002, "epoch": 3.5555555555555554, "step": 5440}, {"loss": 1.2224, "grad_norm": 0.9495334029197693, "learning_rate": 0.0002, "epoch": 3.5620915032679736, "step": 5450}, {"loss": 1.2869, "grad_norm": 0.8743821978569031, "learning_rate": 0.0002, "epoch": 3.568627450980392, "step": 5460}, {"loss": 1.1018, "grad_norm": 0.7513086795806885, "learning_rate": 0.0002, "epoch": 3.57516339869281, "step": 5470}, {"loss": 1.1082, "grad_norm": 1.0139480829238892, "learning_rate": 0.0002, "epoch": 3.581699346405229, "step": 5480}, {"loss": 1.1706, "grad_norm": 0.6615135073661804, "learning_rate": 0.0002, "epoch": 3.588235294117647, "step": 5490}, {"loss": 1.3906, "grad_norm": 1.180798888206482, "learning_rate": 0.0002, "epoch": 3.5947712418300655, "step": 5500}, {"loss": 1.2391, "grad_norm": 0.7085279226303101, "learning_rate": 0.0002, "epoch": 3.6013071895424837, "step": 5510}, {"loss": 1.1623, "grad_norm": 0.540268063545227, "learning_rate": 0.0002, "epoch": 3.607843137254902, "step": 5520}, {"loss": 1.2132, "grad_norm": 0.7905671000480652, "learning_rate": 0.0002, "epoch": 3.6143790849673203, "step": 5530}, {"loss": 1.2731, "grad_norm": 0.8457717299461365, "learning_rate": 0.0002, "epoch": 3.6209150326797386, "step": 5540}, {"loss": 1.1799, "grad_norm": 0.7102677822113037, "learning_rate": 0.0002, "epoch": 3.627450980392157, "step": 5550}, {"loss": 1.2394, "grad_norm": 0.7179514765739441, "learning_rate": 0.0002, "epoch": 3.633986928104575, "step": 5560}, {"loss": 1.2019, "grad_norm": 1.0854148864746094, "learning_rate": 0.0002, "epoch": 3.6405228758169934, "step": 5570}, {"loss": 1.1986, "grad_norm": 0.8209951519966125, "learning_rate": 0.0002, "epoch": 3.6470588235294117, "step": 5580}, {"loss": 1.2289, "grad_norm": 0.6944138407707214, "learning_rate": 0.0002, "epoch": 3.65359477124183, "step": 5590}, {"loss": 1.3226, "grad_norm": 0.7675473093986511, "learning_rate": 0.0002, "epoch": 3.6601307189542482, "step": 5600}, {"loss": 1.2866, "grad_norm": 0.6683364510536194, "learning_rate": 0.0002, "epoch": 3.6666666666666665, "step": 5610}, {"loss": 1.1099, "grad_norm": 0.7920727133750916, "learning_rate": 0.0002, "epoch": 3.6732026143790852, "step": 5620}, {"loss": 1.2287, "grad_norm": 0.9440218806266785, "learning_rate": 0.0002, "epoch": 3.6797385620915035, "step": 5630}, {"loss": 1.2444, "grad_norm": 0.6600824594497681, "learning_rate": 0.0002, "epoch": 3.686274509803922, "step": 5640}, {"loss": 1.191, "grad_norm": 0.6860619187355042, "learning_rate": 0.0002, "epoch": 3.69281045751634, "step": 5650}, {"loss": 1.1914, "grad_norm": 0.6579713225364685, "learning_rate": 0.0002, "epoch": 3.6993464052287583, "step": 5660}, {"loss": 1.1464, "grad_norm": 0.661081075668335, "learning_rate": 0.0002, "epoch": 3.7058823529411766, "step": 5670}, {"loss": 1.289, "grad_norm": 1.0968825817108154, "learning_rate": 0.0002, "epoch": 3.712418300653595, "step": 5680}, {"loss": 1.192, "grad_norm": 0.8066844940185547, "learning_rate": 0.0002, "epoch": 3.718954248366013, "step": 5690}, {"loss": 1.2322, "grad_norm": 0.8341682553291321, "learning_rate": 0.0002, "epoch": 3.7254901960784315, "step": 5700}, {"loss": 1.1473, "grad_norm": 0.6682852506637573, "learning_rate": 0.0002, "epoch": 3.7320261437908497, "step": 5710}, {"loss": 1.1566, "grad_norm": 0.898595929145813, "learning_rate": 0.0002, "epoch": 3.738562091503268, "step": 5720}, {"loss": 1.0919, "grad_norm": 0.6876054406166077, "learning_rate": 0.0002, "epoch": 3.7450980392156863, "step": 5730}, {"loss": 1.2302, "grad_norm": 0.7817103266716003, "learning_rate": 0.0002, "epoch": 3.7516339869281046, "step": 5740}, {"loss": 1.2439, "grad_norm": 0.5840168595314026, "learning_rate": 0.0002, "epoch": 3.758169934640523, "step": 5750}, {"loss": 1.1279, "grad_norm": 0.6263918876647949, "learning_rate": 0.0002, "epoch": 3.764705882352941, "step": 5760}, {"loss": 1.2023, "grad_norm": 0.7948952317237854, "learning_rate": 0.0002, "epoch": 3.7712418300653594, "step": 5770}, {"loss": 1.149, "grad_norm": 0.6700998544692993, "learning_rate": 0.0002, "epoch": 3.7777777777777777, "step": 5780}, {"loss": 1.3207, "grad_norm": 1.1169519424438477, "learning_rate": 0.0002, "epoch": 3.784313725490196, "step": 5790}, {"loss": 1.064, "grad_norm": 0.8354471325874329, "learning_rate": 0.0002, "epoch": 3.7908496732026142, "step": 5800}, {"loss": 1.2104, "grad_norm": 0.6304181814193726, "learning_rate": 0.0002, "epoch": 3.7973856209150325, "step": 5810}, {"loss": 1.2059, "grad_norm": 0.6919655799865723, "learning_rate": 0.0002, "epoch": 3.803921568627451, "step": 5820}, {"loss": 1.217, "grad_norm": 0.600385844707489, "learning_rate": 0.0002, "epoch": 3.810457516339869, "step": 5830}, {"loss": 1.2324, "grad_norm": 0.8406319618225098, "learning_rate": 0.0002, "epoch": 3.8169934640522873, "step": 5840}, {"loss": 1.2418, "grad_norm": 0.7594282031059265, "learning_rate": 0.0002, "epoch": 3.8235294117647056, "step": 5850}, {"loss": 1.1903, "grad_norm": 0.8179879784584045, "learning_rate": 0.0002, "epoch": 3.8300653594771243, "step": 5860}, {"loss": 1.255, "grad_norm": 1.141430377960205, "learning_rate": 0.0002, "epoch": 3.8366013071895426, "step": 5870}, {"loss": 1.1467, "grad_norm": 0.6595550775527954, "learning_rate": 0.0002, "epoch": 3.843137254901961, "step": 5880}, {"loss": 1.2378, "grad_norm": 0.7499435544013977, "learning_rate": 0.0002, "epoch": 3.849673202614379, "step": 5890}, {"loss": 1.217, "grad_norm": 0.7851517200469971, "learning_rate": 0.0002, "epoch": 3.8562091503267975, "step": 5900}, {"loss": 1.162, "grad_norm": 1.0533545017242432, "learning_rate": 0.0002, "epoch": 3.8627450980392157, "step": 5910}, {"loss": 1.3576, "grad_norm": 0.960086464881897, "learning_rate": 0.0002, "epoch": 3.869281045751634, "step": 5920}, {"loss": 1.151, "grad_norm": 0.9952049851417542, "learning_rate": 0.0002, "epoch": 3.8758169934640523, "step": 5930}, {"loss": 1.2027, "grad_norm": 0.7884191274642944, "learning_rate": 0.0002, "epoch": 3.8823529411764706, "step": 5940}, {"loss": 1.1796, "grad_norm": 0.7461766600608826, "learning_rate": 0.0002, "epoch": 3.888888888888889, "step": 5950}, {"loss": 1.2251, "grad_norm": 0.9594355821609497, "learning_rate": 0.0002, "epoch": 3.895424836601307, "step": 5960}, {"loss": 1.1164, "grad_norm": 0.8179471492767334, "learning_rate": 0.0002, "epoch": 3.9019607843137254, "step": 5970}, {"loss": 1.2421, "grad_norm": 0.8240267634391785, "learning_rate": 0.0002, "epoch": 3.9084967320261437, "step": 5980}, {"loss": 1.3076, "grad_norm": 0.7462618350982666, "learning_rate": 0.0002, "epoch": 3.915032679738562, "step": 5990}, {"loss": 1.2124, "grad_norm": 0.711207389831543, "learning_rate": 0.0002, "epoch": 3.9215686274509802, "step": 6000}, {"loss": 1.2119, "grad_norm": 0.6910956501960754, "learning_rate": 0.0002, "epoch": 3.928104575163399, "step": 6010}, {"loss": 1.2127, "grad_norm": 0.749093770980835, "learning_rate": 0.0002, "epoch": 3.9346405228758172, "step": 6020}, {"loss": 1.1542, "grad_norm": 1.3332762718200684, "learning_rate": 0.0002, "epoch": 3.9411764705882355, "step": 6030}, {"loss": 1.1442, "grad_norm": 0.71457439661026, "learning_rate": 0.0002, "epoch": 3.947712418300654, "step": 6040}, {"loss": 1.339, "grad_norm": 1.1205238103866577, "learning_rate": 0.0002, "epoch": 3.954248366013072, "step": 6050}, {"loss": 1.2962, "grad_norm": 0.6958928108215332, "learning_rate": 0.0002, "epoch": 3.9607843137254903, "step": 6060}, {"loss": 1.1802, "grad_norm": 0.7518056035041809, "learning_rate": 0.0002, "epoch": 3.9673202614379086, "step": 6070}, {"loss": 1.1179, "grad_norm": 0.8010755777359009, "learning_rate": 0.0002, "epoch": 3.973856209150327, "step": 6080}, {"loss": 1.2867, "grad_norm": 0.7492658495903015, "learning_rate": 0.0002, "epoch": 3.980392156862745, "step": 6090}, {"loss": 1.2113, "grad_norm": 0.900704562664032, "learning_rate": 0.0002, "epoch": 3.9869281045751634, "step": 6100}, {"loss": 1.1106, "grad_norm": 0.7997331619262695, "learning_rate": 0.0002, "epoch": 3.9934640522875817, "step": 6110}, {"loss": 1.1244, "grad_norm": 0.7163209319114685, "learning_rate": 0.0002, "epoch": 4.0, "step": 6120}, {"eval_loss": 1.4113320112228394, "eval_runtime": 33.7199, "eval_samples_per_second": 12.93, "eval_steps_per_second": 1.631, "epoch": 4.0, "step": 6120}, {"loss": 1.0423, "grad_norm": 0.9527022838592529, "learning_rate": 0.0002, "epoch": 4.006535947712418, "step": 6130}, {"loss": 1.101, "grad_norm": 0.7603210210800171, "learning_rate": 0.0002, "epoch": 4.0130718954248366, "step": 6140}, {"loss": 1.1834, "grad_norm": 1.127387523651123, "learning_rate": 0.0002, "epoch": 4.019607843137255, "step": 6150}, {"loss": 1.0734, "grad_norm": 0.8290133476257324, "learning_rate": 0.0002, "epoch": 4.026143790849673, "step": 6160}, {"loss": 1.0785, "grad_norm": 0.9912241101264954, "learning_rate": 0.0002, "epoch": 4.032679738562091, "step": 6170}, {"loss": 1.0719, "grad_norm": 0.947005033493042, "learning_rate": 0.0002, "epoch": 4.03921568627451, "step": 6180}, {"loss": 1.0835, "grad_norm": 0.707466185092926, "learning_rate": 0.0002, "epoch": 4.045751633986928, "step": 6190}, {"loss": 1.1079, "grad_norm": 1.0604327917099, "learning_rate": 0.0002, "epoch": 4.052287581699346, "step": 6200}, {"loss": 1.0375, "grad_norm": 0.7848685383796692, "learning_rate": 0.0002, "epoch": 4.0588235294117645, "step": 6210}, {"loss": 1.1167, "grad_norm": 0.8475256562232971, "learning_rate": 0.0002, "epoch": 4.065359477124183, "step": 6220}, {"loss": 1.1104, "grad_norm": 0.9759448766708374, "learning_rate": 0.0002, "epoch": 4.071895424836601, "step": 6230}, {"loss": 1.1538, "grad_norm": 0.9324519038200378, "learning_rate": 0.0002, "epoch": 4.078431372549019, "step": 6240}, {"loss": 1.0817, "grad_norm": 0.8723901510238647, "learning_rate": 0.0002, "epoch": 4.084967320261438, "step": 6250}, {"loss": 1.0977, "grad_norm": 0.8343415856361389, "learning_rate": 0.0002, "epoch": 4.091503267973856, "step": 6260}, {"loss": 0.9887, "grad_norm": 0.7490310072898865, "learning_rate": 0.0002, "epoch": 4.098039215686274, "step": 6270}, {"loss": 1.2084, "grad_norm": 0.8961182832717896, "learning_rate": 0.0002, "epoch": 4.104575163398692, "step": 6280}, {"loss": 1.1349, "grad_norm": 0.7124854922294617, "learning_rate": 0.0002, "epoch": 4.111111111111111, "step": 6290}, {"loss": 1.0081, "grad_norm": 0.8338138461112976, "learning_rate": 0.0002, "epoch": 4.117647058823529, "step": 6300}, {"loss": 1.1091, "grad_norm": 0.8075833320617676, "learning_rate": 0.0002, "epoch": 4.124183006535947, "step": 6310}, {"loss": 1.0193, "grad_norm": 0.8069391846656799, "learning_rate": 0.0002, "epoch": 4.130718954248366, "step": 6320}, {"loss": 0.948, "grad_norm": 0.9567893147468567, "learning_rate": 0.0002, "epoch": 4.137254901960785, "step": 6330}, {"loss": 1.0241, "grad_norm": 1.2184662818908691, "learning_rate": 0.0002, "epoch": 4.143790849673203, "step": 6340}, {"loss": 1.0756, "grad_norm": 1.030976414680481, "learning_rate": 0.0002, "epoch": 4.150326797385621, "step": 6350}, {"loss": 1.1124, "grad_norm": 0.9749957323074341, "learning_rate": 0.0002, "epoch": 4.1568627450980395, "step": 6360}, {"loss": 1.1038, "grad_norm": 0.7089483141899109, "learning_rate": 0.0002, "epoch": 4.163398692810458, "step": 6370}, {"loss": 1.2175, "grad_norm": 1.1084946393966675, "learning_rate": 0.0002, "epoch": 4.169934640522876, "step": 6380}, {"loss": 1.0274, "grad_norm": 0.7998497486114502, "learning_rate": 0.0002, "epoch": 4.176470588235294, "step": 6390}, {"loss": 1.005, "grad_norm": 0.8997811675071716, "learning_rate": 0.0002, "epoch": 4.183006535947713, "step": 6400}, {"loss": 1.0704, "grad_norm": 0.8359479904174805, "learning_rate": 0.0002, "epoch": 4.189542483660131, "step": 6410}, {"loss": 1.1056, "grad_norm": 0.9087472558021545, "learning_rate": 0.0002, "epoch": 4.196078431372549, "step": 6420}, {"loss": 1.0657, "grad_norm": 1.1100451946258545, "learning_rate": 0.0002, "epoch": 4.2026143790849675, "step": 6430}, {"loss": 1.1443, "grad_norm": 0.9376999735832214, "learning_rate": 0.0002, "epoch": 4.209150326797386, "step": 6440}, {"loss": 1.0862, "grad_norm": 0.8179266452789307, "learning_rate": 0.0002, "epoch": 4.215686274509804, "step": 6450}, {"loss": 1.0679, "grad_norm": 0.9953271746635437, "learning_rate": 0.0002, "epoch": 4.222222222222222, "step": 6460}, {"loss": 1.1034, "grad_norm": 0.8476650714874268, "learning_rate": 0.0002, "epoch": 4.228758169934641, "step": 6470}, {"loss": 1.2512, "grad_norm": 0.8406323194503784, "learning_rate": 0.0002, "epoch": 4.235294117647059, "step": 6480}, {"loss": 1.057, "grad_norm": 0.819134533405304, "learning_rate": 0.0002, "epoch": 4.241830065359477, "step": 6490}, {"loss": 1.1082, "grad_norm": 0.7764983773231506, "learning_rate": 0.0002, "epoch": 4.248366013071895, "step": 6500}, {"loss": 1.1593, "grad_norm": 0.8252112865447998, "learning_rate": 0.0002, "epoch": 4.254901960784314, "step": 6510}, {"loss": 1.1369, "grad_norm": 0.7941019535064697, "learning_rate": 0.0002, "epoch": 4.261437908496732, "step": 6520}, {"loss": 1.0296, "grad_norm": 0.7673905491828918, "learning_rate": 0.0002, "epoch": 4.26797385620915, "step": 6530}, {"loss": 1.1387, "grad_norm": 0.8749890327453613, "learning_rate": 0.0002, "epoch": 4.2745098039215685, "step": 6540}, {"loss": 1.0595, "grad_norm": 0.7343207597732544, "learning_rate": 0.0002, "epoch": 4.281045751633987, "step": 6550}, {"loss": 1.1715, "grad_norm": 1.2786651849746704, "learning_rate": 0.0002, "epoch": 4.287581699346405, "step": 6560}, {"loss": 1.0514, "grad_norm": 1.316875696182251, "learning_rate": 0.0002, "epoch": 4.294117647058823, "step": 6570}, {"loss": 1.1125, "grad_norm": 0.8349189162254333, "learning_rate": 0.0002, "epoch": 4.300653594771242, "step": 6580}, {"loss": 1.0732, "grad_norm": 0.7510647177696228, "learning_rate": 0.0002, "epoch": 4.30718954248366, "step": 6590}, {"loss": 1.1387, "grad_norm": 0.932420551776886, "learning_rate": 0.0002, "epoch": 4.313725490196078, "step": 6600}, {"loss": 1.1115, "grad_norm": 0.8510616421699524, "learning_rate": 0.0002, "epoch": 4.3202614379084965, "step": 6610}, {"loss": 1.0957, "grad_norm": 0.7661547064781189, "learning_rate": 0.0002, "epoch": 4.326797385620915, "step": 6620}, {"loss": 1.2064, "grad_norm": 1.0370930433273315, "learning_rate": 0.0002, "epoch": 4.333333333333333, "step": 6630}, {"loss": 1.1064, "grad_norm": 0.9302158951759338, "learning_rate": 0.0002, "epoch": 4.339869281045751, "step": 6640}, {"loss": 0.968, "grad_norm": 0.9203811883926392, "learning_rate": 0.0002, "epoch": 4.34640522875817, "step": 6650}, {"loss": 1.0123, "grad_norm": 0.9986332654953003, "learning_rate": 0.0002, "epoch": 4.352941176470588, "step": 6660}, {"loss": 1.1079, "grad_norm": 0.8001713156700134, "learning_rate": 0.0002, "epoch": 4.359477124183006, "step": 6670}, {"loss": 1.0248, "grad_norm": 0.829714298248291, "learning_rate": 0.0002, "epoch": 4.366013071895424, "step": 6680}, {"loss": 1.0389, "grad_norm": 0.8253079056739807, "learning_rate": 0.0002, "epoch": 4.372549019607844, "step": 6690}, {"loss": 1.1087, "grad_norm": 0.824666440486908, "learning_rate": 0.0002, "epoch": 4.379084967320262, "step": 6700}, {"loss": 1.1968, "grad_norm": 0.8872972130775452, "learning_rate": 0.0002, "epoch": 4.38562091503268, "step": 6710}, {"loss": 1.0474, "grad_norm": 0.8729761838912964, "learning_rate": 0.0002, "epoch": 4.392156862745098, "step": 6720}, {"loss": 1.0961, "grad_norm": 1.1367264986038208, "learning_rate": 0.0002, "epoch": 4.398692810457517, "step": 6730}, {"loss": 1.0184, "grad_norm": 0.9699058532714844, "learning_rate": 0.0002, "epoch": 4.405228758169935, "step": 6740}, {"loss": 1.006, "grad_norm": 0.8266763687133789, "learning_rate": 0.0002, "epoch": 4.411764705882353, "step": 6750}, {"loss": 1.0735, "grad_norm": 1.0249767303466797, "learning_rate": 0.0002, "epoch": 4.4183006535947715, "step": 6760}, {"loss": 1.1726, "grad_norm": 0.73606938123703, "learning_rate": 0.0002, "epoch": 4.42483660130719, "step": 6770}, {"loss": 1.1037, "grad_norm": 1.4050679206848145, "learning_rate": 0.0002, "epoch": 4.431372549019608, "step": 6780}, {"loss": 1.1418, "grad_norm": 1.1114081144332886, "learning_rate": 0.0002, "epoch": 4.437908496732026, "step": 6790}, {"loss": 0.9682, "grad_norm": 0.8031067848205566, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6800}, {"loss": 1.0753, "grad_norm": 0.8513566851615906, "learning_rate": 0.0002, "epoch": 4.450980392156863, "step": 6810}, {"loss": 1.1852, "grad_norm": 1.332741379737854, "learning_rate": 0.0002, "epoch": 4.457516339869281, "step": 6820}, {"loss": 1.0966, "grad_norm": 1.5032578706741333, "learning_rate": 0.0002, "epoch": 4.4640522875816995, "step": 6830}, {"loss": 1.1124, "grad_norm": 0.7677283883094788, "learning_rate": 0.0002, "epoch": 4.470588235294118, "step": 6840}, {"loss": 1.1501, "grad_norm": 0.989148736000061, "learning_rate": 0.0002, "epoch": 4.477124183006536, "step": 6850}, {"loss": 1.2239, "grad_norm": 1.5316275358200073, "learning_rate": 0.0002, "epoch": 4.483660130718954, "step": 6860}, {"loss": 1.1171, "grad_norm": 0.9427124261856079, "learning_rate": 0.0002, "epoch": 4.490196078431373, "step": 6870}, {"loss": 1.1314, "grad_norm": 1.215287685394287, "learning_rate": 0.0002, "epoch": 4.496732026143791, "step": 6880}, {"loss": 1.0809, "grad_norm": 0.7286760210990906, "learning_rate": 0.0002, "epoch": 4.503267973856209, "step": 6890}, {"loss": 1.0179, "grad_norm": 0.874829888343811, "learning_rate": 0.0002, "epoch": 4.509803921568627, "step": 6900}, {"loss": 1.0233, "grad_norm": 0.8058359622955322, "learning_rate": 0.0002, "epoch": 4.516339869281046, "step": 6910}, {"loss": 1.0463, "grad_norm": 1.248195767402649, "learning_rate": 0.0002, "epoch": 4.522875816993464, "step": 6920}, {"loss": 1.0347, "grad_norm": 0.8033645749092102, "learning_rate": 0.0002, "epoch": 4.529411764705882, "step": 6930}, {"loss": 1.1068, "grad_norm": 1.7361950874328613, "learning_rate": 0.0002, "epoch": 4.5359477124183005, "step": 6940}, {"loss": 0.9856, "grad_norm": 0.8058095574378967, "learning_rate": 0.0002, "epoch": 4.542483660130719, "step": 6950}, {"loss": 1.0057, "grad_norm": 1.254089593887329, "learning_rate": 0.0002, "epoch": 4.549019607843137, "step": 6960}, {"loss": 1.1723, "grad_norm": 0.9180455803871155, "learning_rate": 0.0002, "epoch": 4.555555555555555, "step": 6970}, {"loss": 1.0559, "grad_norm": 0.6677682399749756, "learning_rate": 0.0002, "epoch": 4.562091503267974, "step": 6980}, {"loss": 1.0453, "grad_norm": 0.8127354383468628, "learning_rate": 0.0002, "epoch": 4.568627450980392, "step": 6990}, {"loss": 1.0828, "grad_norm": 1.0263001918792725, "learning_rate": 0.0002, "epoch": 4.57516339869281, "step": 7000}, {"loss": 1.0703, "grad_norm": 0.9641909003257751, "learning_rate": 0.0002, "epoch": 4.5816993464052285, "step": 7010}, {"loss": 1.179, "grad_norm": 0.9440861344337463, "learning_rate": 0.0002, "epoch": 4.588235294117647, "step": 7020}, {"loss": 1.0931, "grad_norm": 0.9539011716842651, "learning_rate": 0.0002, "epoch": 4.594771241830065, "step": 7030}, {"loss": 1.0963, "grad_norm": 1.0449910163879395, "learning_rate": 0.0002, "epoch": 4.601307189542483, "step": 7040}, {"loss": 0.9944, "grad_norm": 0.8766893744468689, "learning_rate": 0.0002, "epoch": 4.607843137254902, "step": 7050}, {"loss": 1.0169, "grad_norm": 0.6983462572097778, "learning_rate": 0.0002, "epoch": 4.61437908496732, "step": 7060}, {"loss": 1.1778, "grad_norm": 0.9505505561828613, "learning_rate": 0.0002, "epoch": 4.620915032679738, "step": 7070}, {"loss": 1.121, "grad_norm": 1.2506657838821411, "learning_rate": 0.0002, "epoch": 4.627450980392156, "step": 7080}, {"loss": 1.1329, "grad_norm": 0.9602801203727722, "learning_rate": 0.0002, "epoch": 4.633986928104575, "step": 7090}, {"loss": 1.1499, "grad_norm": 0.7398977875709534, "learning_rate": 0.0002, "epoch": 4.640522875816993, "step": 7100}, {"loss": 1.0769, "grad_norm": 1.3862425088882446, "learning_rate": 0.0002, "epoch": 4.647058823529412, "step": 7110}, {"loss": 1.0571, "grad_norm": 1.1451990604400635, "learning_rate": 0.0002, "epoch": 4.65359477124183, "step": 7120}, {"loss": 1.1271, "grad_norm": 0.9010422229766846, "learning_rate": 0.0002, "epoch": 4.660130718954249, "step": 7130}, {"loss": 1.0165, "grad_norm": 0.7102518081665039, "learning_rate": 0.0002, "epoch": 4.666666666666667, "step": 7140}, {"loss": 1.0819, "grad_norm": 0.7963796257972717, "learning_rate": 0.0002, "epoch": 4.673202614379085, "step": 7150}, {"loss": 1.1114, "grad_norm": 0.7726007699966431, "learning_rate": 0.0002, "epoch": 4.6797385620915035, "step": 7160}, {"loss": 1.2088, "grad_norm": 0.8097564578056335, "learning_rate": 0.0002, "epoch": 4.686274509803922, "step": 7170}, {"loss": 1.1386, "grad_norm": 0.9070925116539001, "learning_rate": 0.0002, "epoch": 4.69281045751634, "step": 7180}, {"loss": 1.0315, "grad_norm": 0.7543528079986572, "learning_rate": 0.0002, "epoch": 4.699346405228758, "step": 7190}, {"loss": 1.0984, "grad_norm": 0.9900904893875122, "learning_rate": 0.0002, "epoch": 4.705882352941177, "step": 7200}, {"loss": 1.1552, "grad_norm": 0.8033412098884583, "learning_rate": 0.0002, "epoch": 4.712418300653595, "step": 7210}, {"loss": 1.1773, "grad_norm": 0.8440839052200317, "learning_rate": 0.0002, "epoch": 4.718954248366013, "step": 7220}, {"loss": 1.1258, "grad_norm": 0.9325555562973022, "learning_rate": 0.0002, "epoch": 4.7254901960784315, "step": 7230}, {"loss": 1.1384, "grad_norm": 0.7881146669387817, "learning_rate": 0.0002, "epoch": 4.73202614379085, "step": 7240}, {"loss": 1.1219, "grad_norm": 0.884453296661377, "learning_rate": 0.0002, "epoch": 4.738562091503268, "step": 7250}, {"loss": 1.1036, "grad_norm": 0.9274539351463318, "learning_rate": 0.0002, "epoch": 4.745098039215686, "step": 7260}, {"loss": 1.0906, "grad_norm": 1.2367479801177979, "learning_rate": 0.0002, "epoch": 4.751633986928105, "step": 7270}, {"loss": 1.0741, "grad_norm": 0.9499821066856384, "learning_rate": 0.0002, "epoch": 4.758169934640523, "step": 7280}, {"loss": 1.1625, "grad_norm": 2.1918580532073975, "learning_rate": 0.0002, "epoch": 4.764705882352941, "step": 7290}, {"loss": 0.954, "grad_norm": 0.8221880793571472, "learning_rate": 0.0002, "epoch": 4.771241830065359, "step": 7300}, {"loss": 1.1358, "grad_norm": 0.871972918510437, "learning_rate": 0.0002, "epoch": 4.777777777777778, "step": 7310}, {"loss": 1.0599, "grad_norm": 0.8034510612487793, "learning_rate": 0.0002, "epoch": 4.784313725490196, "step": 7320}, {"loss": 1.1059, "grad_norm": 0.8959605693817139, "learning_rate": 0.0002, "epoch": 4.790849673202614, "step": 7330}, {"loss": 1.0176, "grad_norm": 1.2326215505599976, "learning_rate": 0.0002, "epoch": 4.7973856209150325, "step": 7340}, {"loss": 1.1095, "grad_norm": 0.9725791811943054, "learning_rate": 0.0002, "epoch": 4.803921568627451, "step": 7350}, {"loss": 1.1229, "grad_norm": 0.7240816354751587, "learning_rate": 0.0002, "epoch": 4.810457516339869, "step": 7360}, {"loss": 1.0669, "grad_norm": 0.8265769481658936, "learning_rate": 0.0002, "epoch": 4.816993464052287, "step": 7370}, {"loss": 1.042, "grad_norm": 0.8888696432113647, "learning_rate": 0.0002, "epoch": 4.823529411764706, "step": 7380}, {"loss": 1.0981, "grad_norm": 0.7776556015014648, "learning_rate": 0.0002, "epoch": 4.830065359477124, "step": 7390}, {"loss": 1.0819, "grad_norm": 0.8772371411323547, "learning_rate": 0.0002, "epoch": 4.836601307189542, "step": 7400}, {"loss": 1.0819, "grad_norm": 0.9786531925201416, "learning_rate": 0.0002, "epoch": 4.8431372549019605, "step": 7410}, {"loss": 1.1358, "grad_norm": 0.9059745073318481, "learning_rate": 0.0002, "epoch": 4.849673202614379, "step": 7420}, {"loss": 1.0324, "grad_norm": 0.7422552108764648, "learning_rate": 0.0002, "epoch": 4.856209150326797, "step": 7430}, {"loss": 1.0423, "grad_norm": 1.3040380477905273, "learning_rate": 0.0002, "epoch": 4.862745098039216, "step": 7440}, {"loss": 1.1161, "grad_norm": 1.3278473615646362, "learning_rate": 0.0002, "epoch": 4.8692810457516345, "step": 7450}, {"loss": 1.0713, "grad_norm": 1.2705849409103394, "learning_rate": 0.0002, "epoch": 4.875816993464053, "step": 7460}, {"loss": 1.0034, "grad_norm": 0.8837892413139343, "learning_rate": 0.0002, "epoch": 4.882352941176471, "step": 7470}, {"loss": 1.1716, "grad_norm": 0.8670691251754761, "learning_rate": 0.0002, "epoch": 4.888888888888889, "step": 7480}, {"loss": 1.1723, "grad_norm": 0.9662758111953735, "learning_rate": 0.0002, "epoch": 4.895424836601308, "step": 7490}, {"loss": 1.1056, "grad_norm": 0.8188302516937256, "learning_rate": 0.0002, "epoch": 4.901960784313726, "step": 7500}, {"loss": 1.0419, "grad_norm": 0.769442617893219, "learning_rate": 0.0002, "epoch": 4.908496732026144, "step": 7510}, {"loss": 1.1671, "grad_norm": 1.1465084552764893, "learning_rate": 0.0002, "epoch": 4.915032679738562, "step": 7520}, {"loss": 1.0768, "grad_norm": 1.253214955329895, "learning_rate": 0.0002, "epoch": 4.921568627450981, "step": 7530}, {"loss": 1.011, "grad_norm": 0.7922375202178955, "learning_rate": 0.0002, "epoch": 4.928104575163399, "step": 7540}, {"loss": 1.1256, "grad_norm": 0.8306851387023926, "learning_rate": 0.0002, "epoch": 4.934640522875817, "step": 7550}, {"loss": 1.206, "grad_norm": 0.8486151099205017, "learning_rate": 0.0002, "epoch": 4.9411764705882355, "step": 7560}, {"loss": 1.0161, "grad_norm": 1.2601467370986938, "learning_rate": 0.0002, "epoch": 4.947712418300654, "step": 7570}, {"loss": 1.1078, "grad_norm": 0.7980747818946838, "learning_rate": 0.0002, "epoch": 4.954248366013072, "step": 7580}, {"loss": 1.0607, "grad_norm": 0.8653254508972168, "learning_rate": 0.0002, "epoch": 4.96078431372549, "step": 7590}, {"loss": 1.0292, "grad_norm": 0.9680571556091309, "learning_rate": 0.0002, "epoch": 4.967320261437909, "step": 7600}, {"loss": 1.1795, "grad_norm": 0.9554466605186462, "learning_rate": 0.0002, "epoch": 4.973856209150327, "step": 7610}, {"loss": 1.0935, "grad_norm": 1.3693897724151611, "learning_rate": 0.0002, "epoch": 4.980392156862745, "step": 7620}, {"loss": 1.0838, "grad_norm": 0.7809282541275024, "learning_rate": 0.0002, "epoch": 4.9869281045751634, "step": 7630}, {"loss": 1.0844, "grad_norm": 0.7528006434440613, "learning_rate": 0.0002, "epoch": 4.993464052287582, "step": 7640}, {"loss": 0.9951, "grad_norm": 1.7491309642791748, "learning_rate": 0.0002, "epoch": 5.0, "step": 7650}, {"eval_loss": 1.4197258949279785, "eval_runtime": 33.6327, "eval_samples_per_second": 12.964, "eval_steps_per_second": 1.635, "epoch": 5.0, "step": 7650}, {"loss": 0.9744, "grad_norm": 0.8840063214302063, "learning_rate": 0.0002, "epoch": 5.006535947712418, "step": 7660}, {"loss": 1.0274, "grad_norm": 1.0118401050567627, "learning_rate": 0.0002, "epoch": 5.0130718954248366, "step": 7670}, {"loss": 1.1667, "grad_norm": 1.0040518045425415, "learning_rate": 0.0002, "epoch": 5.019607843137255, "step": 7680}, {"loss": 0.9426, "grad_norm": 0.7541199922561646, "learning_rate": 0.0002, "epoch": 5.026143790849673, "step": 7690}, {"loss": 1.0797, "grad_norm": 0.9106482863426208, "learning_rate": 0.0002, "epoch": 5.032679738562091, "step": 7700}, {"loss": 1.0096, "grad_norm": 1.3691469430923462, "learning_rate": 0.0002, "epoch": 5.03921568627451, "step": 7710}, {"loss": 0.9889, "grad_norm": 0.9449689388275146, "learning_rate": 0.0002, "epoch": 5.045751633986928, "step": 7720}, {"loss": 0.9087, "grad_norm": 1.1678508520126343, "learning_rate": 0.0002, "epoch": 5.052287581699346, "step": 7730}, {"loss": 1.0556, "grad_norm": 1.1296145915985107, "learning_rate": 0.0002, "epoch": 5.0588235294117645, "step": 7740}, {"loss": 0.9339, "grad_norm": 0.7863904237747192, "learning_rate": 0.0002, "epoch": 5.065359477124183, "step": 7750}, {"loss": 1.0135, "grad_norm": 0.8691433072090149, "learning_rate": 0.0002, "epoch": 5.071895424836601, "step": 7760}, {"loss": 0.9776, "grad_norm": 1.0722088813781738, "learning_rate": 0.0002, "epoch": 5.078431372549019, "step": 7770}, {"loss": 1.0595, "grad_norm": 0.9625038504600525, "learning_rate": 0.0002, "epoch": 5.084967320261438, "step": 7780}, {"loss": 1.0241, "grad_norm": 1.2618783712387085, "learning_rate": 0.0002, "epoch": 5.091503267973856, "step": 7790}, {"loss": 0.9396, "grad_norm": 0.9970650672912598, "learning_rate": 0.0002, "epoch": 5.098039215686274, "step": 7800}, {"loss": 0.9186, "grad_norm": 1.3946677446365356, "learning_rate": 0.0002, "epoch": 5.104575163398692, "step": 7810}, {"loss": 0.9957, "grad_norm": 1.0260052680969238, "learning_rate": 0.0002, "epoch": 5.111111111111111, "step": 7820}, {"loss": 0.9865, "grad_norm": 1.105521559715271, "learning_rate": 0.0002, "epoch": 5.117647058823529, "step": 7830}, {"loss": 0.9788, "grad_norm": 1.003641128540039, "learning_rate": 0.0002, "epoch": 5.124183006535947, "step": 7840}, {"loss": 0.9688, "grad_norm": 1.0315021276474, "learning_rate": 0.0002, "epoch": 5.130718954248366, "step": 7850}, {"loss": 1.0001, "grad_norm": 0.9469530582427979, "learning_rate": 0.0002, "epoch": 5.137254901960785, "step": 7860}, {"loss": 0.9659, "grad_norm": 1.3244667053222656, "learning_rate": 0.0002, "epoch": 5.143790849673203, "step": 7870}, {"loss": 0.9657, "grad_norm": 1.1732033491134644, "learning_rate": 0.0002, "epoch": 5.150326797385621, "step": 7880}, {"loss": 0.9978, "grad_norm": 1.3129149675369263, "learning_rate": 0.0002, "epoch": 5.1568627450980395, "step": 7890}, {"loss": 0.9894, "grad_norm": 0.8589454293251038, "learning_rate": 0.0002, "epoch": 5.163398692810458, "step": 7900}, {"loss": 1.0161, "grad_norm": 0.8954233527183533, "learning_rate": 0.0002, "epoch": 5.169934640522876, "step": 7910}, {"loss": 0.8741, "grad_norm": 0.7426522970199585, "learning_rate": 0.0002, "epoch": 5.176470588235294, "step": 7920}, {"loss": 1.0106, "grad_norm": 1.1990121603012085, "learning_rate": 0.0002, "epoch": 5.183006535947713, "step": 7930}, {"loss": 0.9453, "grad_norm": 0.8867580890655518, "learning_rate": 0.0002, "epoch": 5.189542483660131, "step": 7940}, {"loss": 0.9727, "grad_norm": 1.016276478767395, "learning_rate": 0.0002, "epoch": 5.196078431372549, "step": 7950}, {"loss": 0.9908, "grad_norm": 1.0210685729980469, "learning_rate": 0.0002, "epoch": 5.2026143790849675, "step": 7960}, {"loss": 1.0522, "grad_norm": 1.0093122720718384, "learning_rate": 0.0002, "epoch": 5.209150326797386, "step": 7970}, {"loss": 1.0055, "grad_norm": 0.9746801853179932, "learning_rate": 0.0002, "epoch": 5.215686274509804, "step": 7980}, {"loss": 1.0611, "grad_norm": 0.9113537073135376, "learning_rate": 0.0002, "epoch": 5.222222222222222, "step": 7990}, {"loss": 0.9167, "grad_norm": 1.2782206535339355, "learning_rate": 0.0002, "epoch": 5.228758169934641, "step": 8000}, {"loss": 1.0212, "grad_norm": 1.3223118782043457, "learning_rate": 0.0002, "epoch": 5.235294117647059, "step": 8010}, {"loss": 0.9244, "grad_norm": 0.7898629307746887, "learning_rate": 0.0002, "epoch": 5.241830065359477, "step": 8020}, {"loss": 1.0574, "grad_norm": 0.9822350740432739, "learning_rate": 0.0002, "epoch": 5.248366013071895, "step": 8030}, {"loss": 1.0102, "grad_norm": 1.5114340782165527, "learning_rate": 0.0002, "epoch": 5.254901960784314, "step": 8040}, {"loss": 0.9816, "grad_norm": 0.859006941318512, "learning_rate": 0.0002, "epoch": 5.261437908496732, "step": 8050}, {"loss": 0.9445, "grad_norm": 1.0495043992996216, "learning_rate": 0.0002, "epoch": 5.26797385620915, "step": 8060}, {"loss": 0.9724, "grad_norm": 1.329483151435852, "learning_rate": 0.0002, "epoch": 5.2745098039215685, "step": 8070}, {"loss": 0.9296, "grad_norm": 1.1333061456680298, "learning_rate": 0.0002, "epoch": 5.281045751633987, "step": 8080}, {"loss": 0.9577, "grad_norm": 0.8153108358383179, "learning_rate": 0.0002, "epoch": 5.287581699346405, "step": 8090}, {"loss": 0.9002, "grad_norm": 0.9395004510879517, "learning_rate": 0.0002, "epoch": 5.294117647058823, "step": 8100}, {"loss": 1.0371, "grad_norm": 0.8907593488693237, "learning_rate": 0.0002, "epoch": 5.300653594771242, "step": 8110}, {"loss": 0.9301, "grad_norm": 0.9808667898178101, "learning_rate": 0.0002, "epoch": 5.30718954248366, "step": 8120}, {"loss": 1.0136, "grad_norm": 0.984779417514801, "learning_rate": 0.0002, "epoch": 5.313725490196078, "step": 8130}, {"loss": 0.9621, "grad_norm": 0.9787270426750183, "learning_rate": 0.0002, "epoch": 5.3202614379084965, "step": 8140}, {"loss": 0.9336, "grad_norm": 0.9857710599899292, "learning_rate": 0.0002, "epoch": 5.326797385620915, "step": 8150}, {"loss": 0.9884, "grad_norm": 0.9774303436279297, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 8160}, {"loss": 1.0561, "grad_norm": 0.677925169467926, "learning_rate": 0.0002, "epoch": 5.339869281045751, "step": 8170}, {"loss": 1.1345, "grad_norm": 0.9576456546783447, "learning_rate": 0.0002, "epoch": 5.34640522875817, "step": 8180}, {"loss": 0.9554, "grad_norm": 1.8970937728881836, "learning_rate": 0.0002, "epoch": 5.352941176470588, "step": 8190}, {"loss": 1.0474, "grad_norm": 0.9458389282226562, "learning_rate": 0.0002, "epoch": 5.359477124183006, "step": 8200}, {"loss": 1.0365, "grad_norm": 1.761794924736023, "learning_rate": 0.0002, "epoch": 5.366013071895424, "step": 8210}, {"loss": 0.9426, "grad_norm": 1.0693724155426025, "learning_rate": 0.0002, "epoch": 5.372549019607844, "step": 8220}, {"loss": 1.0299, "grad_norm": 0.9025877714157104, "learning_rate": 0.0002, "epoch": 5.379084967320262, "step": 8230}, {"loss": 0.9652, "grad_norm": 1.258857250213623, "learning_rate": 0.0002, "epoch": 5.38562091503268, "step": 8240}, {"loss": 0.9735, "grad_norm": 1.084849238395691, "learning_rate": 0.0002, "epoch": 5.392156862745098, "step": 8250}, {"loss": 0.9999, "grad_norm": 0.9530340433120728, "learning_rate": 0.0002, "epoch": 5.398692810457517, "step": 8260}, {"loss": 1.0268, "grad_norm": 0.830240786075592, "learning_rate": 0.0002, "epoch": 5.405228758169935, "step": 8270}, {"loss": 1.0332, "grad_norm": 1.5807015895843506, "learning_rate": 0.0002, "epoch": 5.411764705882353, "step": 8280}, {"loss": 0.9146, "grad_norm": 0.9486905336380005, "learning_rate": 0.0002, "epoch": 5.4183006535947715, "step": 8290}, {"loss": 1.0336, "grad_norm": 1.0415093898773193, "learning_rate": 0.0002, "epoch": 5.42483660130719, "step": 8300}, {"loss": 0.8933, "grad_norm": 1.0501102209091187, "learning_rate": 0.0002, "epoch": 5.431372549019608, "step": 8310}, {"loss": 0.9983, "grad_norm": 0.9751836061477661, "learning_rate": 0.0002, "epoch": 5.437908496732026, "step": 8320}, {"loss": 1.0755, "grad_norm": 1.5529173612594604, "learning_rate": 0.0002, "epoch": 5.444444444444445, "step": 8330}, {"loss": 0.9814, "grad_norm": 0.8314350247383118, "learning_rate": 0.0002, "epoch": 5.450980392156863, "step": 8340}, {"loss": 1.0596, "grad_norm": 1.2555103302001953, "learning_rate": 0.0002, "epoch": 5.457516339869281, "step": 8350}, {"loss": 1.0127, "grad_norm": 0.9408367872238159, "learning_rate": 0.0002, "epoch": 5.4640522875816995, "step": 8360}, {"loss": 0.9241, "grad_norm": 0.9483312964439392, "learning_rate": 0.0002, "epoch": 5.470588235294118, "step": 8370}, {"loss": 0.9678, "grad_norm": 0.957905650138855, "learning_rate": 0.0002, "epoch": 5.477124183006536, "step": 8380}, {"loss": 1.0985, "grad_norm": 1.4000147581100464, "learning_rate": 0.0002, "epoch": 5.483660130718954, "step": 8390}, {"loss": 0.9966, "grad_norm": 1.7032461166381836, "learning_rate": 0.0002, "epoch": 5.490196078431373, "step": 8400}, {"loss": 0.9539, "grad_norm": 0.8978716731071472, "learning_rate": 0.0002, "epoch": 5.496732026143791, "step": 8410}, {"loss": 0.9544, "grad_norm": 0.8659300804138184, "learning_rate": 0.0002, "epoch": 5.503267973856209, "step": 8420}, {"loss": 1.0526, "grad_norm": 1.3629727363586426, "learning_rate": 0.0002, "epoch": 5.509803921568627, "step": 8430}, {"loss": 0.9696, "grad_norm": 1.2741984128952026, "learning_rate": 0.0002, "epoch": 5.516339869281046, "step": 8440}, {"loss": 1.0191, "grad_norm": 1.3867180347442627, "learning_rate": 0.0002, "epoch": 5.522875816993464, "step": 8450}, {"loss": 1.0835, "grad_norm": 1.0662001371383667, "learning_rate": 0.0002, "epoch": 5.529411764705882, "step": 8460}, {"loss": 0.9779, "grad_norm": 1.7005380392074585, "learning_rate": 0.0002, "epoch": 5.5359477124183005, "step": 8470}, {"loss": 1.0221, "grad_norm": 1.3730385303497314, "learning_rate": 0.0002, "epoch": 5.542483660130719, "step": 8480}, {"loss": 0.9586, "grad_norm": 1.7737441062927246, "learning_rate": 0.0002, "epoch": 5.549019607843137, "step": 8490}, {"loss": 0.9729, "grad_norm": 0.907487690448761, "learning_rate": 0.0002, "epoch": 5.555555555555555, "step": 8500}, {"loss": 0.9891, "grad_norm": 0.8882441520690918, "learning_rate": 0.0002, "epoch": 5.562091503267974, "step": 8510}, {"loss": 0.973, "grad_norm": 0.8655388951301575, "learning_rate": 0.0002, "epoch": 5.568627450980392, "step": 8520}, {"loss": 0.9523, "grad_norm": 1.379992961883545, "learning_rate": 0.0002, "epoch": 5.57516339869281, "step": 8530}, {"loss": 1.0174, "grad_norm": 1.0021201372146606, "learning_rate": 0.0002, "epoch": 5.5816993464052285, "step": 8540}, {"loss": 1.0113, "grad_norm": 1.2636926174163818, "learning_rate": 0.0002, "epoch": 5.588235294117647, "step": 8550}, {"loss": 1.0243, "grad_norm": 1.279025912284851, "learning_rate": 0.0002, "epoch": 5.594771241830065, "step": 8560}, {"loss": 0.9917, "grad_norm": 0.8885834217071533, "learning_rate": 0.0002, "epoch": 5.601307189542483, "step": 8570}, {"loss": 0.9849, "grad_norm": 1.1975032091140747, "learning_rate": 0.0002, "epoch": 5.607843137254902, "step": 8580}, {"loss": 1.0363, "grad_norm": 1.005470871925354, "learning_rate": 0.0002, "epoch": 5.61437908496732, "step": 8590}, {"loss": 0.9947, "grad_norm": 1.104286551475525, "learning_rate": 0.0002, "epoch": 5.620915032679738, "step": 8600}, {"loss": 1.0585, "grad_norm": 1.435445785522461, "learning_rate": 0.0002, "epoch": 5.627450980392156, "step": 8610}, {"loss": 0.9156, "grad_norm": 1.0270172357559204, "learning_rate": 0.0002, "epoch": 5.633986928104575, "step": 8620}, {"loss": 1.0522, "grad_norm": 1.0929527282714844, "learning_rate": 0.0002, "epoch": 5.640522875816993, "step": 8630}, {"loss": 0.9694, "grad_norm": 1.1061221361160278, "learning_rate": 0.0002, "epoch": 5.647058823529412, "step": 8640}, {"loss": 1.0826, "grad_norm": 0.9563149213790894, "learning_rate": 0.0002, "epoch": 5.65359477124183, "step": 8650}, {"loss": 1.0042, "grad_norm": 1.0434954166412354, "learning_rate": 0.0002, "epoch": 5.660130718954249, "step": 8660}, {"loss": 0.9463, "grad_norm": 1.3695117235183716, "learning_rate": 0.0002, "epoch": 5.666666666666667, "step": 8670}, {"loss": 0.9441, "grad_norm": 1.0540564060211182, "learning_rate": 0.0002, "epoch": 5.673202614379085, "step": 8680}, {"loss": 0.9755, "grad_norm": 1.5942492485046387, "learning_rate": 0.0002, "epoch": 5.6797385620915035, "step": 8690}, {"loss": 1.0071, "grad_norm": 0.9485495090484619, "learning_rate": 0.0002, "epoch": 5.686274509803922, "step": 8700}, {"loss": 0.9998, "grad_norm": 1.1483162641525269, "learning_rate": 0.0002, "epoch": 5.69281045751634, "step": 8710}, {"loss": 0.9578, "grad_norm": 0.9075471758842468, "learning_rate": 0.0002, "epoch": 5.699346405228758, "step": 8720}, {"loss": 0.9488, "grad_norm": 1.7908551692962646, "learning_rate": 0.0002, "epoch": 5.705882352941177, "step": 8730}, {"loss": 1.0163, "grad_norm": 0.8867162466049194, "learning_rate": 0.0002, "epoch": 5.712418300653595, "step": 8740}, {"loss": 1.0041, "grad_norm": 1.7165148258209229, "learning_rate": 0.0002, "epoch": 5.718954248366013, "step": 8750}, {"loss": 1.1061, "grad_norm": 0.9529356956481934, "learning_rate": 0.0002, "epoch": 5.7254901960784315, "step": 8760}, {"loss": 1.1119, "grad_norm": 1.01852548122406, "learning_rate": 0.0002, "epoch": 5.73202614379085, "step": 8770}, {"loss": 1.0471, "grad_norm": 0.9538423418998718, "learning_rate": 0.0002, "epoch": 5.738562091503268, "step": 8780}, {"loss": 1.0913, "grad_norm": 0.9007737636566162, "learning_rate": 0.0002, "epoch": 5.745098039215686, "step": 8790}, {"loss": 0.9766, "grad_norm": 0.9107874035835266, "learning_rate": 0.0002, "epoch": 5.751633986928105, "step": 8800}, {"loss": 0.9212, "grad_norm": 0.7379238605499268, "learning_rate": 0.0002, "epoch": 5.758169934640523, "step": 8810}, {"loss": 1.0966, "grad_norm": 1.072645902633667, "learning_rate": 0.0002, "epoch": 5.764705882352941, "step": 8820}, {"loss": 1.0845, "grad_norm": 1.002008080482483, "learning_rate": 0.0002, "epoch": 5.771241830065359, "step": 8830}, {"loss": 0.9978, "grad_norm": 1.0435924530029297, "learning_rate": 0.0002, "epoch": 5.777777777777778, "step": 8840}, {"loss": 0.9458, "grad_norm": 0.9874551296234131, "learning_rate": 0.0002, "epoch": 5.784313725490196, "step": 8850}, {"loss": 1.1241, "grad_norm": 1.1729662418365479, "learning_rate": 0.0002, "epoch": 5.790849673202614, "step": 8860}, {"loss": 1.0451, "grad_norm": 1.3300775289535522, "learning_rate": 0.0002, "epoch": 5.7973856209150325, "step": 8870}, {"loss": 1.0989, "grad_norm": 1.612707257270813, "learning_rate": 0.0002, "epoch": 5.803921568627451, "step": 8880}, {"loss": 0.9119, "grad_norm": 0.9047797322273254, "learning_rate": 0.0002, "epoch": 5.810457516339869, "step": 8890}, {"loss": 0.989, "grad_norm": 1.0958741903305054, "learning_rate": 0.0002, "epoch": 5.816993464052287, "step": 8900}, {"loss": 1.1922, "grad_norm": 1.0099612474441528, "learning_rate": 0.0002, "epoch": 5.823529411764706, "step": 8910}, {"loss": 1.0623, "grad_norm": 0.8442328572273254, "learning_rate": 0.0002, "epoch": 5.830065359477124, "step": 8920}, {"loss": 0.9134, "grad_norm": 1.1388301849365234, "learning_rate": 0.0002, "epoch": 5.836601307189542, "step": 8930}, {"loss": 1.0019, "grad_norm": 0.8296026587486267, "learning_rate": 0.0002, "epoch": 5.8431372549019605, "step": 8940}, {"loss": 1.0363, "grad_norm": 1.0843533277511597, "learning_rate": 0.0002, "epoch": 5.849673202614379, "step": 8950}, {"loss": 1.0009, "grad_norm": 0.8496834635734558, "learning_rate": 0.0002, "epoch": 5.856209150326797, "step": 8960}, {"loss": 0.9927, "grad_norm": 1.6894690990447998, "learning_rate": 0.0002, "epoch": 5.862745098039216, "step": 8970}, {"loss": 1.0939, "grad_norm": 1.0012282133102417, "learning_rate": 0.0002, "epoch": 5.8692810457516345, "step": 8980}, {"loss": 0.9722, "grad_norm": 0.8521103262901306, "learning_rate": 0.0002, "epoch": 5.875816993464053, "step": 8990}, {"loss": 1.0885, "grad_norm": 1.246841311454773, "learning_rate": 0.0002, "epoch": 5.882352941176471, "step": 9000}, {"loss": 0.9702, "grad_norm": 0.9941892027854919, "learning_rate": 0.0002, "epoch": 5.888888888888889, "step": 9010}, {"loss": 0.8754, "grad_norm": 1.067413568496704, "learning_rate": 0.0002, "epoch": 5.895424836601308, "step": 9020}, {"loss": 1.0153, "grad_norm": 1.0045088529586792, "learning_rate": 0.0002, "epoch": 5.901960784313726, "step": 9030}, {"loss": 1.0134, "grad_norm": 1.383063554763794, "learning_rate": 0.0002, "epoch": 5.908496732026144, "step": 9040}, {"loss": 1.0845, "grad_norm": 0.8754428625106812, "learning_rate": 0.0002, "epoch": 5.915032679738562, "step": 9050}, {"loss": 0.9571, "grad_norm": 0.8577388525009155, "learning_rate": 0.0002, "epoch": 5.921568627450981, "step": 9060}, {"loss": 1.0532, "grad_norm": 0.8718975186347961, "learning_rate": 0.0002, "epoch": 5.928104575163399, "step": 9070}, {"loss": 1.0667, "grad_norm": 1.1762131452560425, "learning_rate": 0.0002, "epoch": 5.934640522875817, "step": 9080}, {"loss": 1.1114, "grad_norm": 1.1025866270065308, "learning_rate": 0.0002, "epoch": 5.9411764705882355, "step": 9090}, {"loss": 0.9155, "grad_norm": 1.0439870357513428, "learning_rate": 0.0002, "epoch": 5.947712418300654, "step": 9100}, {"loss": 1.0055, "grad_norm": 1.2411525249481201, "learning_rate": 0.0002, "epoch": 5.954248366013072, "step": 9110}, {"loss": 0.9747, "grad_norm": 1.0317714214324951, "learning_rate": 0.0002, "epoch": 5.96078431372549, "step": 9120}, {"loss": 1.0352, "grad_norm": 0.9880492091178894, "learning_rate": 0.0002, "epoch": 5.967320261437909, "step": 9130}, {"loss": 1.0459, "grad_norm": 0.9039815664291382, "learning_rate": 0.0002, "epoch": 5.973856209150327, "step": 9140}, {"loss": 1.0413, "grad_norm": 0.9049116373062134, "learning_rate": 0.0002, "epoch": 5.980392156862745, "step": 9150}, {"loss": 0.9792, "grad_norm": 0.996749222278595, "learning_rate": 0.0002, "epoch": 5.9869281045751634, "step": 9160}, {"loss": 0.8857, "grad_norm": 0.8716062307357788, "learning_rate": 0.0002, "epoch": 5.993464052287582, "step": 9170}, {"loss": 1.019, "grad_norm": 1.3081294298171997, "learning_rate": 0.0002, "epoch": 6.0, "step": 9180}, {"eval_loss": 1.45111083984375, "eval_runtime": 34.7121, "eval_samples_per_second": 12.56, "eval_steps_per_second": 1.584, "epoch": 6.0, "step": 9180}, {"loss": 0.9306, "grad_norm": 1.1378029584884644, "learning_rate": 0.0002, "epoch": 6.006535947712418, "step": 9190}, {"loss": 0.8794, "grad_norm": 1.2921233177185059, "learning_rate": 0.0002, "epoch": 6.0130718954248366, "step": 9200}, {"loss": 0.8145, "grad_norm": 1.039211630821228, "learning_rate": 0.0002, "epoch": 6.019607843137255, "step": 9210}, {"loss": 0.8524, "grad_norm": 0.9715196490287781, "learning_rate": 0.0002, "epoch": 6.026143790849673, "step": 9220}, {"loss": 1.035, "grad_norm": 1.220642328262329, "learning_rate": 0.0002, "epoch": 6.032679738562091, "step": 9230}, {"loss": 0.8468, "grad_norm": 0.854360044002533, "learning_rate": 0.0002, "epoch": 6.03921568627451, "step": 9240}, {"loss": 0.8534, "grad_norm": 0.8806933164596558, "learning_rate": 0.0002, "epoch": 6.045751633986928, "step": 9250}, {"loss": 0.8305, "grad_norm": 1.4315874576568604, "learning_rate": 0.0002, "epoch": 6.052287581699346, "step": 9260}, {"loss": 0.8462, "grad_norm": 0.9382007122039795, "learning_rate": 0.0002, "epoch": 6.0588235294117645, "step": 9270}, {"loss": 0.9653, "grad_norm": 1.2184561491012573, "learning_rate": 0.0002, "epoch": 6.065359477124183, "step": 9280}, {"loss": 0.8806, "grad_norm": 1.2331548929214478, "learning_rate": 0.0002, "epoch": 6.071895424836601, "step": 9290}, {"loss": 0.8354, "grad_norm": 1.1112796068191528, "learning_rate": 0.0002, "epoch": 6.078431372549019, "step": 9300}, {"loss": 0.8008, "grad_norm": 1.4753731489181519, "learning_rate": 0.0002, "epoch": 6.084967320261438, "step": 9310}, {"loss": 0.9198, "grad_norm": 1.2783401012420654, "learning_rate": 0.0002, "epoch": 6.091503267973856, "step": 9320}, {"loss": 0.8294, "grad_norm": 0.9916909337043762, "learning_rate": 0.0002, "epoch": 6.098039215686274, "step": 9330}, {"loss": 0.876, "grad_norm": 0.9300099015235901, "learning_rate": 0.0002, "epoch": 6.104575163398692, "step": 9340}, {"loss": 0.9064, "grad_norm": 1.4985264539718628, "learning_rate": 0.0002, "epoch": 6.111111111111111, "step": 9350}, {"loss": 1.0106, "grad_norm": 1.276380181312561, "learning_rate": 0.0002, "epoch": 6.117647058823529, "step": 9360}, {"loss": 0.9068, "grad_norm": 1.181113600730896, "learning_rate": 0.0002, "epoch": 6.124183006535947, "step": 9370}, {"loss": 0.9165, "grad_norm": 1.698729395866394, "learning_rate": 0.0002, "epoch": 6.130718954248366, "step": 9380}, {"loss": 0.7997, "grad_norm": 0.9793189764022827, "learning_rate": 0.0002, "epoch": 6.137254901960785, "step": 9390}, {"loss": 0.9731, "grad_norm": 1.1942132711410522, "learning_rate": 0.0002, "epoch": 6.143790849673203, "step": 9400}, {"loss": 0.8762, "grad_norm": 1.2160184383392334, "learning_rate": 0.0002, "epoch": 6.150326797385621, "step": 9410}, {"loss": 0.801, "grad_norm": 1.0802825689315796, "learning_rate": 0.0002, "epoch": 6.1568627450980395, "step": 9420}, {"loss": 0.9055, "grad_norm": 3.024529218673706, "learning_rate": 0.0002, "epoch": 6.163398692810458, "step": 9430}, {"loss": 0.8739, "grad_norm": 0.975062370300293, "learning_rate": 0.0002, "epoch": 6.169934640522876, "step": 9440}, {"loss": 0.8485, "grad_norm": 0.9243306517601013, "learning_rate": 0.0002, "epoch": 6.176470588235294, "step": 9450}, {"loss": 0.947, "grad_norm": 0.8892099857330322, "learning_rate": 0.0002, "epoch": 6.183006535947713, "step": 9460}, {"loss": 0.9165, "grad_norm": 1.4151731729507446, "learning_rate": 0.0002, "epoch": 6.189542483660131, "step": 9470}, {"loss": 1.022, "grad_norm": 1.064701795578003, "learning_rate": 0.0002, "epoch": 6.196078431372549, "step": 9480}, {"loss": 0.906, "grad_norm": 1.1104519367218018, "learning_rate": 0.0002, "epoch": 6.2026143790849675, "step": 9490}, {"loss": 0.9572, "grad_norm": 1.4788947105407715, "learning_rate": 0.0002, "epoch": 6.209150326797386, "step": 9500}, {"loss": 0.8014, "grad_norm": 0.7976077795028687, "learning_rate": 0.0002, "epoch": 6.215686274509804, "step": 9510}, {"loss": 0.886, "grad_norm": 1.256864070892334, "learning_rate": 0.0002, "epoch": 6.222222222222222, "step": 9520}, {"loss": 0.9104, "grad_norm": 1.3874554634094238, "learning_rate": 0.0002, "epoch": 6.228758169934641, "step": 9530}, {"loss": 0.8583, "grad_norm": 1.9012963771820068, "learning_rate": 0.0002, "epoch": 6.235294117647059, "step": 9540}, {"loss": 0.9585, "grad_norm": 1.275212287902832, "learning_rate": 0.0002, "epoch": 6.241830065359477, "step": 9550}, {"loss": 0.8416, "grad_norm": 1.1007417440414429, "learning_rate": 0.0002, "epoch": 6.248366013071895, "step": 9560}, {"loss": 0.9191, "grad_norm": 1.0602147579193115, "learning_rate": 0.0002, "epoch": 6.254901960784314, "step": 9570}, {"loss": 0.909, "grad_norm": 1.2276418209075928, "learning_rate": 0.0002, "epoch": 6.261437908496732, "step": 9580}, {"loss": 0.9363, "grad_norm": 1.0111924409866333, "learning_rate": 0.0002, "epoch": 6.26797385620915, "step": 9590}, {"loss": 0.9941, "grad_norm": 0.9031485915184021, "learning_rate": 0.0002, "epoch": 6.2745098039215685, "step": 9600}, {"loss": 0.9138, "grad_norm": 0.9893783926963806, "learning_rate": 0.0002, "epoch": 6.281045751633987, "step": 9610}, {"loss": 0.9114, "grad_norm": 1.1979725360870361, "learning_rate": 0.0002, "epoch": 6.287581699346405, "step": 9620}, {"loss": 0.8858, "grad_norm": 1.380516767501831, "learning_rate": 0.0002, "epoch": 6.294117647058823, "step": 9630}, {"loss": 0.8898, "grad_norm": 1.1370083093643188, "learning_rate": 0.0002, "epoch": 6.300653594771242, "step": 9640}, {"loss": 0.9073, "grad_norm": 1.4091558456420898, "learning_rate": 0.0002, "epoch": 6.30718954248366, "step": 9650}, {"loss": 0.9096, "grad_norm": 1.0670944452285767, "learning_rate": 0.0002, "epoch": 6.313725490196078, "step": 9660}, {"loss": 0.9376, "grad_norm": 0.9150263667106628, "learning_rate": 0.0002, "epoch": 6.3202614379084965, "step": 9670}, {"loss": 0.9169, "grad_norm": 1.1342853307724, "learning_rate": 0.0002, "epoch": 6.326797385620915, "step": 9680}, {"loss": 1.002, "grad_norm": 1.2733415365219116, "learning_rate": 0.0002, "epoch": 6.333333333333333, "step": 9690}, {"loss": 0.9579, "grad_norm": 1.3647292852401733, "learning_rate": 0.0002, "epoch": 6.339869281045751, "step": 9700}, {"loss": 0.87, "grad_norm": 1.0435094833374023, "learning_rate": 0.0002, "epoch": 6.34640522875817, "step": 9710}, {"loss": 0.8812, "grad_norm": 1.3641071319580078, "learning_rate": 0.0002, "epoch": 6.352941176470588, "step": 9720}, {"loss": 0.8888, "grad_norm": 1.2806159257888794, "learning_rate": 0.0002, "epoch": 6.359477124183006, "step": 9730}, {"loss": 0.9481, "grad_norm": 1.0193076133728027, "learning_rate": 0.0002, "epoch": 6.366013071895424, "step": 9740}, {"loss": 0.931, "grad_norm": 1.2349408864974976, "learning_rate": 0.0002, "epoch": 6.372549019607844, "step": 9750}, {"loss": 0.8837, "grad_norm": 1.2062549591064453, "learning_rate": 0.0002, "epoch": 6.379084967320262, "step": 9760}, {"loss": 0.8947, "grad_norm": 1.4402194023132324, "learning_rate": 0.0002, "epoch": 6.38562091503268, "step": 9770}, {"loss": 0.8724, "grad_norm": 1.1730891466140747, "learning_rate": 0.0002, "epoch": 6.392156862745098, "step": 9780}, {"loss": 0.9005, "grad_norm": 1.1481093168258667, "learning_rate": 0.0002, "epoch": 6.398692810457517, "step": 9790}, {"loss": 0.9431, "grad_norm": 1.0012723207473755, "learning_rate": 0.0002, "epoch": 6.405228758169935, "step": 9800}, {"loss": 0.8856, "grad_norm": 0.8839848041534424, "learning_rate": 0.0002, "epoch": 6.411764705882353, "step": 9810}, {"loss": 0.8147, "grad_norm": 1.096693992614746, "learning_rate": 0.0002, "epoch": 6.4183006535947715, "step": 9820}, {"loss": 0.846, "grad_norm": 1.4713369607925415, "learning_rate": 0.0002, "epoch": 6.42483660130719, "step": 9830}, {"loss": 0.9563, "grad_norm": 1.2529761791229248, "learning_rate": 0.0002, "epoch": 6.431372549019608, "step": 9840}, {"loss": 0.8551, "grad_norm": 1.5575600862503052, "learning_rate": 0.0002, "epoch": 6.437908496732026, "step": 9850}, {"loss": 0.836, "grad_norm": 1.2188916206359863, "learning_rate": 0.0002, "epoch": 6.444444444444445, "step": 9860}, {"loss": 0.9132, "grad_norm": 1.1558794975280762, "learning_rate": 0.0002, "epoch": 6.450980392156863, "step": 9870}, {"loss": 0.8632, "grad_norm": 1.1506937742233276, "learning_rate": 0.0002, "epoch": 6.457516339869281, "step": 9880}, {"loss": 1.0575, "grad_norm": 1.1168335676193237, "learning_rate": 0.0002, "epoch": 6.4640522875816995, "step": 9890}, {"loss": 0.99, "grad_norm": 1.192449688911438, "learning_rate": 0.0002, "epoch": 6.470588235294118, "step": 9900}, {"loss": 0.9478, "grad_norm": 1.0451104640960693, "learning_rate": 0.0002, "epoch": 6.477124183006536, "step": 9910}, {"loss": 0.9034, "grad_norm": 1.1111775636672974, "learning_rate": 0.0002, "epoch": 6.483660130718954, "step": 9920}, {"loss": 0.8971, "grad_norm": 1.2094531059265137, "learning_rate": 0.0002, "epoch": 6.490196078431373, "step": 9930}, {"loss": 0.9047, "grad_norm": 1.0547380447387695, "learning_rate": 0.0002, "epoch": 6.496732026143791, "step": 9940}, {"loss": 1.0727, "grad_norm": 1.5547202825546265, "learning_rate": 0.0002, "epoch": 6.503267973856209, "step": 9950}, {"loss": 0.9109, "grad_norm": 1.1917903423309326, "learning_rate": 0.0002, "epoch": 6.509803921568627, "step": 9960}, {"loss": 0.8708, "grad_norm": 1.0918153524398804, "learning_rate": 0.0002, "epoch": 6.516339869281046, "step": 9970}, {"loss": 0.8752, "grad_norm": 1.146968960762024, "learning_rate": 0.0002, "epoch": 6.522875816993464, "step": 9980}, {"loss": 0.9593, "grad_norm": 0.9899234771728516, "learning_rate": 0.0002, "epoch": 6.529411764705882, "step": 9990}, {"loss": 0.91, "grad_norm": 2.160924196243286, "learning_rate": 0.0002, "epoch": 6.5359477124183005, "step": 10000}, {"loss": 0.9683, "grad_norm": 1.6366891860961914, "learning_rate": 0.0002, "epoch": 6.542483660130719, "step": 10010}, {"loss": 0.8582, "grad_norm": 0.9876762628555298, "learning_rate": 0.0002, "epoch": 6.549019607843137, "step": 10020}, {"loss": 0.8385, "grad_norm": 1.5622549057006836, "learning_rate": 0.0002, "epoch": 6.555555555555555, "step": 10030}, {"loss": 0.8791, "grad_norm": 1.0108020305633545, "learning_rate": 0.0002, "epoch": 6.562091503267974, "step": 10040}, {"loss": 0.9574, "grad_norm": 1.0725725889205933, "learning_rate": 0.0002, "epoch": 6.568627450980392, "step": 10050}, {"loss": 0.8297, "grad_norm": 1.1551216840744019, "learning_rate": 0.0002, "epoch": 6.57516339869281, "step": 10060}, {"loss": 0.8199, "grad_norm": 1.5174646377563477, "learning_rate": 0.0002, "epoch": 6.5816993464052285, "step": 10070}, {"loss": 0.8203, "grad_norm": 1.041877031326294, "learning_rate": 0.0002, "epoch": 6.588235294117647, "step": 10080}, {"loss": 0.9684, "grad_norm": 0.9939621686935425, "learning_rate": 0.0002, "epoch": 6.594771241830065, "step": 10090}, {"loss": 0.9324, "grad_norm": 1.2706589698791504, "learning_rate": 0.0002, "epoch": 6.601307189542483, "step": 10100}, {"loss": 0.9614, "grad_norm": 1.1071467399597168, "learning_rate": 0.0002, "epoch": 6.607843137254902, "step": 10110}, {"loss": 0.9747, "grad_norm": 0.9449541568756104, "learning_rate": 0.0002, "epoch": 6.61437908496732, "step": 10120}, {"loss": 0.9557, "grad_norm": 1.0961830615997314, "learning_rate": 0.0002, "epoch": 6.620915032679738, "step": 10130}, {"loss": 0.9865, "grad_norm": 1.7726300954818726, "learning_rate": 0.0002, "epoch": 6.627450980392156, "step": 10140}, {"loss": 0.9657, "grad_norm": 1.2345516681671143, "learning_rate": 0.0002, "epoch": 6.633986928104575, "step": 10150}, {"loss": 0.9573, "grad_norm": 1.2062907218933105, "learning_rate": 0.0002, "epoch": 6.640522875816993, "step": 10160}, {"loss": 0.918, "grad_norm": 1.029327154159546, "learning_rate": 0.0002, "epoch": 6.647058823529412, "step": 10170}, {"loss": 0.9211, "grad_norm": 1.442307710647583, "learning_rate": 0.0002, "epoch": 6.65359477124183, "step": 10180}, {"loss": 0.8924, "grad_norm": 1.2579066753387451, "learning_rate": 0.0002, "epoch": 6.660130718954249, "step": 10190}, {"loss": 0.9836, "grad_norm": 1.4563188552856445, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 10200}, {"loss": 0.8876, "grad_norm": 0.9699450135231018, "learning_rate": 0.0002, "epoch": 6.673202614379085, "step": 10210}, {"loss": 0.9589, "grad_norm": 1.812523603439331, "learning_rate": 0.0002, "epoch": 6.6797385620915035, "step": 10220}, {"loss": 1.0241, "grad_norm": 1.124000906944275, "learning_rate": 0.0002, "epoch": 6.686274509803922, "step": 10230}, {"loss": 0.8924, "grad_norm": 1.0957475900650024, "learning_rate": 0.0002, "epoch": 6.69281045751634, "step": 10240}, {"loss": 0.8891, "grad_norm": 0.989689826965332, "learning_rate": 0.0002, "epoch": 6.699346405228758, "step": 10250}, {"loss": 0.9049, "grad_norm": 1.4353317022323608, "learning_rate": 0.0002, "epoch": 6.705882352941177, "step": 10260}, {"loss": 0.9311, "grad_norm": 1.0245451927185059, "learning_rate": 0.0002, "epoch": 6.712418300653595, "step": 10270}, {"loss": 0.8814, "grad_norm": 1.097334861755371, "learning_rate": 0.0002, "epoch": 6.718954248366013, "step": 10280}, {"loss": 0.9927, "grad_norm": 0.982356071472168, "learning_rate": 0.0002, "epoch": 6.7254901960784315, "step": 10290}, {"loss": 0.9909, "grad_norm": 1.8842819929122925, "learning_rate": 0.0002, "epoch": 6.73202614379085, "step": 10300}, {"loss": 0.9286, "grad_norm": 0.8648947477340698, "learning_rate": 0.0002, "epoch": 6.738562091503268, "step": 10310}, {"loss": 0.987, "grad_norm": 1.1510577201843262, "learning_rate": 0.0002, "epoch": 6.745098039215686, "step": 10320}, {"loss": 0.9217, "grad_norm": 1.874495506286621, "learning_rate": 0.0002, "epoch": 6.751633986928105, "step": 10330}, {"loss": 0.8914, "grad_norm": 1.1126408576965332, "learning_rate": 0.0002, "epoch": 6.758169934640523, "step": 10340}, {"loss": 0.8508, "grad_norm": 1.6654644012451172, "learning_rate": 0.0002, "epoch": 6.764705882352941, "step": 10350}, {"loss": 0.9653, "grad_norm": 1.0699580907821655, "learning_rate": 0.0002, "epoch": 6.771241830065359, "step": 10360}, {"loss": 0.882, "grad_norm": 0.9460757374763489, "learning_rate": 0.0002, "epoch": 6.777777777777778, "step": 10370}, {"loss": 0.9589, "grad_norm": 1.2553058862686157, "learning_rate": 0.0002, "epoch": 6.784313725490196, "step": 10380}, {"loss": 0.8782, "grad_norm": 1.0939891338348389, "learning_rate": 0.0002, "epoch": 6.790849673202614, "step": 10390}, {"loss": 0.9189, "grad_norm": 1.0647451877593994, "learning_rate": 0.0002, "epoch": 6.7973856209150325, "step": 10400}, {"loss": 0.9478, "grad_norm": 1.0954521894454956, "learning_rate": 0.0002, "epoch": 6.803921568627451, "step": 10410}, {"loss": 1.0385, "grad_norm": 1.4371392726898193, "learning_rate": 0.0002, "epoch": 6.810457516339869, "step": 10420}, {"loss": 1.0024, "grad_norm": 1.0063464641571045, "learning_rate": 0.0002, "epoch": 6.816993464052287, "step": 10430}, {"loss": 0.8737, "grad_norm": 1.5189263820648193, "learning_rate": 0.0002, "epoch": 6.823529411764706, "step": 10440}, {"loss": 0.9246, "grad_norm": 0.9715501070022583, "learning_rate": 0.0002, "epoch": 6.830065359477124, "step": 10450}, {"loss": 0.9659, "grad_norm": 1.114586353302002, "learning_rate": 0.0002, "epoch": 6.836601307189542, "step": 10460}, {"loss": 1.0081, "grad_norm": 1.2991431951522827, "learning_rate": 0.0002, "epoch": 6.8431372549019605, "step": 10470}, {"loss": 0.9323, "grad_norm": 1.203114628791809, "learning_rate": 0.0002, "epoch": 6.849673202614379, "step": 10480}, {"loss": 1.0032, "grad_norm": 1.476167917251587, "learning_rate": 0.0002, "epoch": 6.856209150326797, "step": 10490}, {"loss": 1.0275, "grad_norm": 1.0933326482772827, "learning_rate": 0.0002, "epoch": 6.862745098039216, "step": 10500}, {"loss": 1.0068, "grad_norm": 1.2831504344940186, "learning_rate": 0.0002, "epoch": 6.8692810457516345, "step": 10510}, {"loss": 0.9973, "grad_norm": 1.1967637538909912, "learning_rate": 0.0002, "epoch": 6.875816993464053, "step": 10520}, {"loss": 0.9549, "grad_norm": 1.1276888847351074, "learning_rate": 0.0002, "epoch": 6.882352941176471, "step": 10530}, {"loss": 0.9568, "grad_norm": 1.2680490016937256, "learning_rate": 0.0002, "epoch": 6.888888888888889, "step": 10540}, {"loss": 0.9177, "grad_norm": 1.5469038486480713, "learning_rate": 0.0002, "epoch": 6.895424836601308, "step": 10550}, {"loss": 0.8545, "grad_norm": 1.1731038093566895, "learning_rate": 0.0002, "epoch": 6.901960784313726, "step": 10560}, {"loss": 0.9795, "grad_norm": 0.968008816242218, "learning_rate": 0.0002, "epoch": 6.908496732026144, "step": 10570}, {"loss": 0.9439, "grad_norm": 0.9082416892051697, "learning_rate": 0.0002, "epoch": 6.915032679738562, "step": 10580}, {"loss": 0.9898, "grad_norm": 1.5816899538040161, "learning_rate": 0.0002, "epoch": 6.921568627450981, "step": 10590}, {"loss": 0.9692, "grad_norm": 0.9462234377861023, "learning_rate": 0.0002, "epoch": 6.928104575163399, "step": 10600}, {"loss": 1.0193, "grad_norm": 1.4950200319290161, "learning_rate": 0.0002, "epoch": 6.934640522875817, "step": 10610}, {"loss": 0.8888, "grad_norm": 1.2929182052612305, "learning_rate": 0.0002, "epoch": 6.9411764705882355, "step": 10620}, {"loss": 1.0141, "grad_norm": 1.2995754480361938, "learning_rate": 0.0002, "epoch": 6.947712418300654, "step": 10630}, {"loss": 0.9863, "grad_norm": 0.9407122135162354, "learning_rate": 0.0002, "epoch": 6.954248366013072, "step": 10640}, {"loss": 0.9041, "grad_norm": 1.1735378503799438, "learning_rate": 0.0002, "epoch": 6.96078431372549, "step": 10650}, {"loss": 0.936, "grad_norm": 0.9937344193458557, "learning_rate": 0.0002, "epoch": 6.967320261437909, "step": 10660}, {"loss": 0.9577, "grad_norm": 1.2498728036880493, "learning_rate": 0.0002, "epoch": 6.973856209150327, "step": 10670}, {"loss": 1.0504, "grad_norm": 1.0513341426849365, "learning_rate": 0.0002, "epoch": 6.980392156862745, "step": 10680}, {"loss": 0.9259, "grad_norm": 1.4611467123031616, "learning_rate": 0.0002, "epoch": 6.9869281045751634, "step": 10690}, {"loss": 0.9779, "grad_norm": 1.2924799919128418, "learning_rate": 0.0002, "epoch": 6.993464052287582, "step": 10700}, {"loss": 0.8953, "grad_norm": 1.2024929523468018, "learning_rate": 0.0002, "epoch": 7.0, "step": 10710}, {"eval_loss": 1.4972445964813232, "eval_runtime": 33.6225, "eval_samples_per_second": 12.967, "eval_steps_per_second": 1.636, "epoch": 7.0, "step": 10710}, {"loss": 0.8111, "grad_norm": 1.1302162408828735, "learning_rate": 0.0002, "epoch": 7.006535947712418, "step": 10720}, {"loss": 0.8584, "grad_norm": 1.2731552124023438, "learning_rate": 0.0002, "epoch": 7.0130718954248366, "step": 10730}, {"loss": 0.782, "grad_norm": 1.2694480419158936, "learning_rate": 0.0002, "epoch": 7.019607843137255, "step": 10740}, {"loss": 0.7621, "grad_norm": 1.1517360210418701, "learning_rate": 0.0002, "epoch": 7.026143790849673, "step": 10750}, {"loss": 0.9149, "grad_norm": 1.3649171590805054, "learning_rate": 0.0002, "epoch": 7.032679738562091, "step": 10760}, {"loss": 0.7411, "grad_norm": 1.1630656719207764, "learning_rate": 0.0002, "epoch": 7.03921568627451, "step": 10770}, {"loss": 0.8514, "grad_norm": 1.2658313512802124, "learning_rate": 0.0002, "epoch": 7.045751633986928, "step": 10780}, {"loss": 0.8125, "grad_norm": 1.5004769563674927, "learning_rate": 0.0002, "epoch": 7.052287581699346, "step": 10790}, {"loss": 0.754, "grad_norm": 1.052678108215332, "learning_rate": 0.0002, "epoch": 7.0588235294117645, "step": 10800}, {"loss": 0.8613, "grad_norm": 1.3461277484893799, "learning_rate": 0.0002, "epoch": 7.065359477124183, "step": 10810}, {"loss": 0.843, "grad_norm": 1.3074769973754883, "learning_rate": 0.0002, "epoch": 7.071895424836601, "step": 10820}, {"loss": 0.7433, "grad_norm": 1.8454785346984863, "learning_rate": 0.0002, "epoch": 7.078431372549019, "step": 10830}, {"loss": 0.7899, "grad_norm": 0.9786653518676758, "learning_rate": 0.0002, "epoch": 7.084967320261438, "step": 10840}, {"loss": 0.7689, "grad_norm": 1.2760838270187378, "learning_rate": 0.0002, "epoch": 7.091503267973856, "step": 10850}, {"loss": 0.7715, "grad_norm": 1.1340841054916382, "learning_rate": 0.0002, "epoch": 7.098039215686274, "step": 10860}, {"loss": 0.8111, "grad_norm": 1.3808159828186035, "learning_rate": 0.0002, "epoch": 7.104575163398692, "step": 10870}, {"loss": 0.8311, "grad_norm": 1.147668719291687, "learning_rate": 0.0002, "epoch": 7.111111111111111, "step": 10880}, {"loss": 0.8408, "grad_norm": 1.3183035850524902, "learning_rate": 0.0002, "epoch": 7.117647058823529, "step": 10890}, {"loss": 0.7801, "grad_norm": 1.3882936239242554, "learning_rate": 0.0002, "epoch": 7.124183006535947, "step": 10900}, {"loss": 0.7745, "grad_norm": 0.9495398998260498, "learning_rate": 0.0002, "epoch": 7.130718954248366, "step": 10910}, {"loss": 0.8438, "grad_norm": 1.3810124397277832, "learning_rate": 0.0002, "epoch": 7.137254901960785, "step": 10920}, {"loss": 0.8028, "grad_norm": 1.563207745552063, "learning_rate": 0.0002, "epoch": 7.143790849673203, "step": 10930}, {"loss": 0.7562, "grad_norm": 1.2633056640625, "learning_rate": 0.0002, "epoch": 7.150326797385621, "step": 10940}, {"loss": 0.8592, "grad_norm": 1.2398860454559326, "learning_rate": 0.0002, "epoch": 7.1568627450980395, "step": 10950}, {"loss": 0.8467, "grad_norm": 1.166763186454773, "learning_rate": 0.0002, "epoch": 7.163398692810458, "step": 10960}, {"loss": 0.8346, "grad_norm": 1.5083234310150146, "learning_rate": 0.0002, "epoch": 7.169934640522876, "step": 10970}, {"loss": 0.8323, "grad_norm": 1.6927601099014282, "learning_rate": 0.0002, "epoch": 7.176470588235294, "step": 10980}, {"loss": 0.7434, "grad_norm": 1.090780258178711, "learning_rate": 0.0002, "epoch": 7.183006535947713, "step": 10990}, {"loss": 0.784, "grad_norm": 1.0077793598175049, "learning_rate": 0.0002, "epoch": 7.189542483660131, "step": 11000}, {"loss": 0.831, "grad_norm": 1.8293051719665527, "learning_rate": 0.0002, "epoch": 7.196078431372549, "step": 11010}, {"loss": 0.7654, "grad_norm": 1.0761457681655884, "learning_rate": 0.0002, "epoch": 7.2026143790849675, "step": 11020}, {"loss": 0.8395, "grad_norm": 1.0681469440460205, "learning_rate": 0.0002, "epoch": 7.209150326797386, "step": 11030}, {"loss": 0.7983, "grad_norm": 1.961199402809143, "learning_rate": 0.0002, "epoch": 7.215686274509804, "step": 11040}, {"loss": 0.8631, "grad_norm": 1.3750165700912476, "learning_rate": 0.0002, "epoch": 7.222222222222222, "step": 11050}, {"loss": 0.7425, "grad_norm": 1.647005319595337, "learning_rate": 0.0002, "epoch": 7.228758169934641, "step": 11060}, {"loss": 0.844, "grad_norm": 1.1073668003082275, "learning_rate": 0.0002, "epoch": 7.235294117647059, "step": 11070}, {"loss": 0.8292, "grad_norm": 1.450289011001587, "learning_rate": 0.0002, "epoch": 7.241830065359477, "step": 11080}, {"loss": 0.8505, "grad_norm": 1.191163420677185, "learning_rate": 0.0002, "epoch": 7.248366013071895, "step": 11090}, {"loss": 0.8331, "grad_norm": 1.6975404024124146, "learning_rate": 0.0002, "epoch": 7.254901960784314, "step": 11100}, {"loss": 0.8189, "grad_norm": 1.159091830253601, "learning_rate": 0.0002, "epoch": 7.261437908496732, "step": 11110}, {"loss": 0.8612, "grad_norm": 0.9952927827835083, "learning_rate": 0.0002, "epoch": 7.26797385620915, "step": 11120}, {"loss": 0.8299, "grad_norm": 1.4122034311294556, "learning_rate": 0.0002, "epoch": 7.2745098039215685, "step": 11130}, {"loss": 0.8274, "grad_norm": 1.4299325942993164, "learning_rate": 0.0002, "epoch": 7.281045751633987, "step": 11140}, {"loss": 0.9358, "grad_norm": 1.26812744140625, "learning_rate": 0.0002, "epoch": 7.287581699346405, "step": 11150}, {"loss": 0.8036, "grad_norm": 1.0740736722946167, "learning_rate": 0.0002, "epoch": 7.294117647058823, "step": 11160}, {"loss": 0.9005, "grad_norm": 1.1293542385101318, "learning_rate": 0.0002, "epoch": 7.300653594771242, "step": 11170}, {"loss": 0.8775, "grad_norm": 1.3161042928695679, "learning_rate": 0.0002, "epoch": 7.30718954248366, "step": 11180}, {"loss": 0.9586, "grad_norm": 1.5637391805648804, "learning_rate": 0.0002, "epoch": 7.313725490196078, "step": 11190}, {"loss": 0.8704, "grad_norm": 1.3164077997207642, "learning_rate": 0.0002, "epoch": 7.3202614379084965, "step": 11200}, {"loss": 0.8552, "grad_norm": 0.9268870949745178, "learning_rate": 0.0002, "epoch": 7.326797385620915, "step": 11210}, {"loss": 0.8719, "grad_norm": 1.164515733718872, "learning_rate": 0.0002, "epoch": 7.333333333333333, "step": 11220}, {"loss": 0.9011, "grad_norm": 1.6878753900527954, "learning_rate": 0.0002, "epoch": 7.339869281045751, "step": 11230}, {"loss": 0.8417, "grad_norm": 1.1870672702789307, "learning_rate": 0.0002, "epoch": 7.34640522875817, "step": 11240}, {"loss": 0.886, "grad_norm": 1.2923716306686401, "learning_rate": 0.0002, "epoch": 7.352941176470588, "step": 11250}, {"loss": 0.7763, "grad_norm": 1.2006791830062866, "learning_rate": 0.0002, "epoch": 7.359477124183006, "step": 11260}, {"loss": 0.7859, "grad_norm": 1.0424097776412964, "learning_rate": 0.0002, "epoch": 7.366013071895424, "step": 11270}, {"loss": 0.9427, "grad_norm": 1.349094033241272, "learning_rate": 0.0002, "epoch": 7.372549019607844, "step": 11280}, {"loss": 0.8384, "grad_norm": 1.4128005504608154, "learning_rate": 0.0002, "epoch": 7.379084967320262, "step": 11290}, {"loss": 0.9219, "grad_norm": 1.3647041320800781, "learning_rate": 0.0002, "epoch": 7.38562091503268, "step": 11300}, {"loss": 0.7865, "grad_norm": 1.0561704635620117, "learning_rate": 0.0002, "epoch": 7.392156862745098, "step": 11310}, {"loss": 0.9151, "grad_norm": 1.2405760288238525, "learning_rate": 0.0002, "epoch": 7.398692810457517, "step": 11320}, {"loss": 0.9108, "grad_norm": 1.0932328701019287, "learning_rate": 0.0002, "epoch": 7.405228758169935, "step": 11330}, {"loss": 0.873, "grad_norm": 1.206778883934021, "learning_rate": 0.0002, "epoch": 7.411764705882353, "step": 11340}, {"loss": 0.9062, "grad_norm": 1.5261255502700806, "learning_rate": 0.0002, "epoch": 7.4183006535947715, "step": 11350}, {"loss": 0.7795, "grad_norm": 1.4928070306777954, "learning_rate": 0.0002, "epoch": 7.42483660130719, "step": 11360}, {"loss": 0.9027, "grad_norm": 1.331190586090088, "learning_rate": 0.0002, "epoch": 7.431372549019608, "step": 11370}, {"loss": 0.8547, "grad_norm": 1.0745981931686401, "learning_rate": 0.0002, "epoch": 7.437908496732026, "step": 11380}, {"loss": 0.7755, "grad_norm": 1.3070036172866821, "learning_rate": 0.0002, "epoch": 7.444444444444445, "step": 11390}, {"loss": 0.7703, "grad_norm": 0.8743805885314941, "learning_rate": 0.0002, "epoch": 7.450980392156863, "step": 11400}, {"loss": 0.9331, "grad_norm": 1.2747994661331177, "learning_rate": 0.0002, "epoch": 7.457516339869281, "step": 11410}, {"loss": 0.8506, "grad_norm": 1.3688995838165283, "learning_rate": 0.0002, "epoch": 7.4640522875816995, "step": 11420}, {"loss": 0.783, "grad_norm": 1.1788195371627808, "learning_rate": 0.0002, "epoch": 7.470588235294118, "step": 11430}, {"loss": 0.8841, "grad_norm": 2.0186705589294434, "learning_rate": 0.0002, "epoch": 7.477124183006536, "step": 11440}, {"loss": 0.9182, "grad_norm": 1.1707696914672852, "learning_rate": 0.0002, "epoch": 7.483660130718954, "step": 11450}, {"loss": 0.9019, "grad_norm": 1.26426100730896, "learning_rate": 0.0002, "epoch": 7.490196078431373, "step": 11460}, {"loss": 0.8114, "grad_norm": 1.2673691511154175, "learning_rate": 0.0002, "epoch": 7.496732026143791, "step": 11470}, {"loss": 0.9716, "grad_norm": 1.038956642150879, "learning_rate": 0.0002, "epoch": 7.503267973856209, "step": 11480}, {"loss": 0.8066, "grad_norm": 1.216252326965332, "learning_rate": 0.0002, "epoch": 7.509803921568627, "step": 11490}, {"loss": 0.8546, "grad_norm": 1.1520167589187622, "learning_rate": 0.0002, "epoch": 7.516339869281046, "step": 11500}, {"loss": 0.8117, "grad_norm": 1.3962451219558716, "learning_rate": 0.0002, "epoch": 7.522875816993464, "step": 11510}, {"loss": 0.8636, "grad_norm": 1.2226953506469727, "learning_rate": 0.0002, "epoch": 7.529411764705882, "step": 11520}, {"loss": 0.8943, "grad_norm": 1.2891474962234497, "learning_rate": 0.0002, "epoch": 7.5359477124183005, "step": 11530}, {"loss": 0.8787, "grad_norm": 1.3372766971588135, "learning_rate": 0.0002, "epoch": 7.542483660130719, "step": 11540}, {"loss": 0.8103, "grad_norm": 1.4196370840072632, "learning_rate": 0.0002, "epoch": 7.549019607843137, "step": 11550}, {"loss": 0.8722, "grad_norm": 1.0041396617889404, "learning_rate": 0.0002, "epoch": 7.555555555555555, "step": 11560}, {"loss": 0.8512, "grad_norm": 1.3470606803894043, "learning_rate": 0.0002, "epoch": 7.562091503267974, "step": 11570}, {"loss": 0.8472, "grad_norm": 1.1738601922988892, "learning_rate": 0.0002, "epoch": 7.568627450980392, "step": 11580}, {"loss": 0.8816, "grad_norm": 1.1629133224487305, "learning_rate": 0.0002, "epoch": 7.57516339869281, "step": 11590}, {"loss": 0.8502, "grad_norm": 1.2859786748886108, "learning_rate": 0.0002, "epoch": 7.5816993464052285, "step": 11600}, {"loss": 0.8356, "grad_norm": 1.429398775100708, "learning_rate": 0.0002, "epoch": 7.588235294117647, "step": 11610}, {"loss": 0.7914, "grad_norm": 1.3300801515579224, "learning_rate": 0.0002, "epoch": 7.594771241830065, "step": 11620}, {"loss": 0.8671, "grad_norm": 1.3261592388153076, "learning_rate": 0.0002, "epoch": 7.601307189542483, "step": 11630}, {"loss": 0.7651, "grad_norm": 1.8779925107955933, "learning_rate": 0.0002, "epoch": 7.607843137254902, "step": 11640}, {"loss": 0.7666, "grad_norm": 1.7839158773422241, "learning_rate": 0.0002, "epoch": 7.61437908496732, "step": 11650}, {"loss": 0.9163, "grad_norm": 1.6469435691833496, "learning_rate": 0.0002, "epoch": 7.620915032679738, "step": 11660}, {"loss": 0.8589, "grad_norm": 1.5416018962860107, "learning_rate": 0.0002, "epoch": 7.627450980392156, "step": 11670}, {"loss": 0.7112, "grad_norm": 1.5173335075378418, "learning_rate": 0.0002, "epoch": 7.633986928104575, "step": 11680}, {"loss": 0.8958, "grad_norm": 1.1372658014297485, "learning_rate": 0.0002, "epoch": 7.640522875816993, "step": 11690}, {"loss": 0.8852, "grad_norm": 1.233030915260315, "learning_rate": 0.0002, "epoch": 7.647058823529412, "step": 11700}, {"loss": 0.9205, "grad_norm": 1.3100069761276245, "learning_rate": 0.0002, "epoch": 7.65359477124183, "step": 11710}, {"loss": 0.8895, "grad_norm": 0.9770023226737976, "learning_rate": 0.0002, "epoch": 7.660130718954249, "step": 11720}, {"loss": 0.8658, "grad_norm": 1.240946650505066, "learning_rate": 0.0002, "epoch": 7.666666666666667, "step": 11730}, {"loss": 0.9527, "grad_norm": 1.444226861000061, "learning_rate": 0.0002, "epoch": 7.673202614379085, "step": 11740}, {"loss": 0.8448, "grad_norm": 1.2667231559753418, "learning_rate": 0.0002, "epoch": 7.6797385620915035, "step": 11750}, {"loss": 0.9195, "grad_norm": 1.340754747390747, "learning_rate": 0.0002, "epoch": 7.686274509803922, "step": 11760}, {"loss": 0.8829, "grad_norm": 1.181988000869751, "learning_rate": 0.0002, "epoch": 7.69281045751634, "step": 11770}, {"loss": 0.8609, "grad_norm": 1.0623301267623901, "learning_rate": 0.0002, "epoch": 7.699346405228758, "step": 11780}, {"loss": 0.7726, "grad_norm": 1.1917353868484497, "learning_rate": 0.0002, "epoch": 7.705882352941177, "step": 11790}, {"loss": 0.8075, "grad_norm": 1.7202110290527344, "learning_rate": 0.0002, "epoch": 7.712418300653595, "step": 11800}, {"loss": 0.8112, "grad_norm": 1.1121439933776855, "learning_rate": 0.0002, "epoch": 7.718954248366013, "step": 11810}, {"loss": 0.7831, "grad_norm": 0.956794261932373, "learning_rate": 0.0002, "epoch": 7.7254901960784315, "step": 11820}, {"loss": 0.9056, "grad_norm": 1.2524380683898926, "learning_rate": 0.0002, "epoch": 7.73202614379085, "step": 11830}, {"loss": 0.8337, "grad_norm": 1.1095308065414429, "learning_rate": 0.0002, "epoch": 7.738562091503268, "step": 11840}, {"loss": 0.8921, "grad_norm": 1.631195068359375, "learning_rate": 0.0002, "epoch": 7.745098039215686, "step": 11850}, {"loss": 0.8493, "grad_norm": 1.2265965938568115, "learning_rate": 0.0002, "epoch": 7.751633986928105, "step": 11860}, {"loss": 0.8875, "grad_norm": 1.080328106880188, "learning_rate": 0.0002, "epoch": 7.758169934640523, "step": 11870}, {"loss": 0.8732, "grad_norm": 1.5570356845855713, "learning_rate": 0.0002, "epoch": 7.764705882352941, "step": 11880}, {"loss": 0.8293, "grad_norm": 1.3791661262512207, "learning_rate": 0.0002, "epoch": 7.771241830065359, "step": 11890}, {"loss": 0.8333, "grad_norm": 1.1457891464233398, "learning_rate": 0.0002, "epoch": 7.777777777777778, "step": 11900}, {"loss": 0.8546, "grad_norm": 1.6357585191726685, "learning_rate": 0.0002, "epoch": 7.784313725490196, "step": 11910}, {"loss": 0.9041, "grad_norm": 1.1845953464508057, "learning_rate": 0.0002, "epoch": 7.790849673202614, "step": 11920}, {"loss": 0.8364, "grad_norm": 1.2255016565322876, "learning_rate": 0.0002, "epoch": 7.7973856209150325, "step": 11930}, {"loss": 0.913, "grad_norm": 1.2113513946533203, "learning_rate": 0.0002, "epoch": 7.803921568627451, "step": 11940}, {"loss": 0.8553, "grad_norm": 1.0834609270095825, "learning_rate": 0.0002, "epoch": 7.810457516339869, "step": 11950}, {"loss": 0.8125, "grad_norm": 1.0127689838409424, "learning_rate": 0.0002, "epoch": 7.816993464052287, "step": 11960}, {"loss": 0.9468, "grad_norm": 1.1124000549316406, "learning_rate": 0.0002, "epoch": 7.823529411764706, "step": 11970}, {"loss": 0.8345, "grad_norm": 1.3440804481506348, "learning_rate": 0.0002, "epoch": 7.830065359477124, "step": 11980}, {"loss": 0.8939, "grad_norm": 1.8478741645812988, "learning_rate": 0.0002, "epoch": 7.836601307189542, "step": 11990}, {"loss": 0.9708, "grad_norm": 1.1202499866485596, "learning_rate": 0.0002, "epoch": 7.8431372549019605, "step": 12000}, {"loss": 0.8437, "grad_norm": 1.735700249671936, "learning_rate": 0.0002, "epoch": 7.849673202614379, "step": 12010}, {"loss": 0.8333, "grad_norm": 1.2994014024734497, "learning_rate": 0.0002, "epoch": 7.856209150326797, "step": 12020}, {"loss": 0.8656, "grad_norm": 1.8655444383621216, "learning_rate": 0.0002, "epoch": 7.862745098039216, "step": 12030}, {"loss": 0.8919, "grad_norm": 1.0460877418518066, "learning_rate": 0.0002, "epoch": 7.8692810457516345, "step": 12040}, {"loss": 0.8603, "grad_norm": 1.5241339206695557, "learning_rate": 0.0002, "epoch": 7.875816993464053, "step": 12050}, {"loss": 0.8179, "grad_norm": 1.171849250793457, "learning_rate": 0.0002, "epoch": 7.882352941176471, "step": 12060}, {"loss": 0.8577, "grad_norm": 1.0957022905349731, "learning_rate": 0.0002, "epoch": 7.888888888888889, "step": 12070}, {"loss": 0.9212, "grad_norm": 1.4121248722076416, "learning_rate": 0.0002, "epoch": 7.895424836601308, "step": 12080}, {"loss": 1.0002, "grad_norm": 1.3393208980560303, "learning_rate": 0.0002, "epoch": 7.901960784313726, "step": 12090}, {"loss": 0.8959, "grad_norm": 1.1252245903015137, "learning_rate": 0.0002, "epoch": 7.908496732026144, "step": 12100}, {"loss": 0.8494, "grad_norm": 1.4131813049316406, "learning_rate": 0.0002, "epoch": 7.915032679738562, "step": 12110}, {"loss": 0.9106, "grad_norm": 1.2392992973327637, "learning_rate": 0.0002, "epoch": 7.921568627450981, "step": 12120}, {"loss": 0.7951, "grad_norm": 1.3233672380447388, "learning_rate": 0.0002, "epoch": 7.928104575163399, "step": 12130}, {"loss": 0.7853, "grad_norm": 1.2547026872634888, "learning_rate": 0.0002, "epoch": 7.934640522875817, "step": 12140}, {"loss": 0.9203, "grad_norm": 1.1143239736557007, "learning_rate": 0.0002, "epoch": 7.9411764705882355, "step": 12150}, {"loss": 0.8059, "grad_norm": 1.030006766319275, "learning_rate": 0.0002, "epoch": 7.947712418300654, "step": 12160}, {"loss": 0.8076, "grad_norm": 1.1070104837417603, "learning_rate": 0.0002, "epoch": 7.954248366013072, "step": 12170}, {"loss": 0.8191, "grad_norm": 1.3011643886566162, "learning_rate": 0.0002, "epoch": 7.96078431372549, "step": 12180}, {"loss": 0.7951, "grad_norm": 1.134848713874817, "learning_rate": 0.0002, "epoch": 7.967320261437909, "step": 12190}, {"loss": 0.9318, "grad_norm": 1.7021794319152832, "learning_rate": 0.0002, "epoch": 7.973856209150327, "step": 12200}, {"loss": 0.9159, "grad_norm": 1.0190330743789673, "learning_rate": 0.0002, "epoch": 7.980392156862745, "step": 12210}, {"loss": 0.9586, "grad_norm": 1.6083006858825684, "learning_rate": 0.0002, "epoch": 7.9869281045751634, "step": 12220}, {"loss": 0.915, "grad_norm": 0.8929536938667297, "learning_rate": 0.0002, "epoch": 7.993464052287582, "step": 12230}, {"loss": 0.8706, "grad_norm": 0.9928004145622253, "learning_rate": 0.0002, "epoch": 8.0, "step": 12240}]}